III Taxation of Imports

Vito Tanzi, M. Yücelik, Peter Griffith, and Carlos Aguirre
Published Date:
October 1981
  • ShareShare
Show Summary Details


Most countries in the region still rely heavily on the taxation of imports for revenue; this reliance reflects the inadequacies of other tax bases and the countries’ reluctance to make use of other taxes that may require more sophisticated administrative capabilities.1 The consequence has usually been the placement of excessive emphasis on the revenue function of tariffs.

If tariffs are used to provide protection for domestic industries, their effects on economic efficiency should be underscored. Economic development can be greatly hampered by protecting domestic industries to the point of rendering them uneconomic or economically inefficient. Ideally, an important objective of each country should be to develop efficient industries that can make maximum use of its comparative advantages.

As might be expected, there are many remnants of the recent colonial past in most countries of the sub-Saharan region. They appear in the structure of levies, the pattern of trade preferences, the affiliation with regional groups, and the characteristics of national legislation—all of which have been influenced by the colonial history of each country and, in some cases, have been a deterrent to the formulation of national and regional policies.

Other features of the taxation of imports in sub-Saharan African countries are the prevalence of exemptions, the fairly extensive use of specific as opposed to ad valorem rates, and the reliance on price lists rather than on internationally accepted definitions of value for purposes of valuing imports.

Salient Policy Aspects

Tariff Functions

Customs tariffs in the sub-Saharan African countries are reasonably adequate instruments for equity and revenue purposes. However, these positive characteristics have been achieved at the expense of their suitability for protecting domestic production. Because tariffs were originally designed to yield revenue and to improve equity and because practically no country in the region has been able to do away with their revenue function, the tariffs have not been transformed into successful tools of protective policies. At the root of the problem lies a basic dilemma: the maximum revenue cannot be collected if a tariff constructed to afford protection to domestic industries succeeds in its main objective and import substitution takes place.

Moreover, even the equity objective may be inconsistent with revenue considerations; an increase in taxes on luxury products may result in less revenue being collected by deterring consumption, encouraging local production, or encouraging contraband. For example, imports of luxury products in Somalia are subject to combined tax rates ranging from 129 per cent to 736 per cent, the latter rate being applicable to spirits and liquors; taxation at these levels does not normally produce much revenue.

Some limited degree of fine tuning, allowing for a tariff with multiple functions, is possible. The difficulties inherent in doing this, however, dictate a different solution. From the view-point of tax policy, tariffs are just part of the system of indirect taxation. It seems best, therefore, to try and rationalize the latter by assigning a specific goal to each tax instrument. This approach would facilitate considerably the study of the effects that tax changes may have and would make it easier to achieve multiple policy goals.

Under such a rationalized system, customs duties would be used mainly for protection.2 A general sales tax could be the main instrument for collecting revenue, while considerations of equity, energy conservation, and the need to discourage consumption of socially undesirable goods and services would best be served by a system of excise taxes. The general sales tax and the excise taxes should be levied on both imports and domestic production. A clear distinction between import duties and the sales and excise taxes is desirable, even when the bulk of the sales tax base consists of imports and when some of the luxury goods that are to be specially taxed are imported, because the separation avoids giving unintended protection to actual or potential local production.3

Nominal and Effective Protection

The mixing of revenue and protective functions has also led to excessive levels of protection, resulting in damaging effects on resource allocation. Thus, when customs duties have been assessed heavily on luxury goods, an incentive has been created to produce the goods domestically, and it is likely that this kind of import substitution has not improved resource allocation. Recognition of the problem is difficult since effective protection—a concept developed only in the last 15 years—determines the effect of a particular tariff structure on allocative efficiency. While nominal protection levels can be easily measured, rates of effective protection are determined by the value added by local manufacturers, and this is difficult to quantify.4 Nevertheless, it is clear that, because many raw materials, intermediate goods, and capital goods are either tax exempt or taxed at much lower rates than final goods, effective protection of import substitutes—including luxuries—has become excessive. This is particularly true for industries such as batteries, cosmetics, and assembly plant industries in general, where the nominal tariff on the final product is high and the share of domestic value added tends to be low. The seriousness of the problem is underscored by the fact that the costs of this inefficient allocation are likely to rise with the level of industrialization. On the other hand, domestic production of essential food items may have actually been discouraged by negative effective protection resulting from low or zero nominal duties and high domestic value added. Instead of having encouraged production of items in which the countries have noticeable comparative advantages, external economies can be effected, and dependence on foreign exchange availability is lessened, it is quite possible that the protective policies more commonly found have done just the opposite.

Solutions to these problems of protection are not easy to identify. The first step toward solving them should be to undertake a detailed study of the existing protection levels for various industries and sectors. This should be followed by careful determination of the levels of protection that would be compatible with improved resource allocation, efficient utilization of foreign exchange, and appropriate incentives to agriculture. Only then would it be possible to restructure the customs tariff with a view to achieving these objectives. The complexity of this task only highlights the vital importance, pointed out before, of not trying to use the tariff for too many purposes.


It has frequently been observed that the share of items such as machinery and transport equipment in total imports grows as economies develop. Since these items are generally subject to low rates of duty and sometimes are even zero rated, their increased importance vis-á-vis other imports tends to make customs revenue inelastic. By way of illustration, 25 per cent of Sierra Leone’s imports in fiscal year 1977/78 were zero rated; the share of exempt imports in total imports was about the same in Mauritius in 1978. A much higher level of exemptions, about 65 per cent of total imports, was observed in Liberia, which has a higher per capita income, in the period 1973–76. Other more developed economies in the region, such as Cameroon, Gabon, and Senegal, also exempt large shares of their imports.

Fast growth of exemptions has also eroded the bases of import taxes. In many countries, imports needed to carry out government contracts and developmental projects, imports of raw materials, direct government imports, imports of plant and machinery for local manufacturers, imports of agricultural inputs and machinery, and imports of certain quasi-government institutions are partially or totally exempt from duties or other levies. For example, in Sierra Leone, the proportion of the value of imports exempt from the entry fee and the licensing fee went from 36 per cent in fiscal year 1976/77 to 54 per cent of imports in fiscal year 1979/80. Equally, exemptions in Somalia expanded from less than half of the total value of imports in 1977 to more than three fifths the following year. Perhaps the most dangerous institutional arrangement, which is found in some sub-Saharan African countries, is the one conferring broad powers on ministries of finance to grant ad hoc exemptions. When there has been a departure from the principle that legislation on exemptions should be restrictive, considerable pressure has often been exerted on ministry staff to grant more exemptions, leading to a proliferation of exemptions. These pressures, however, may have had their origin in excessive duties which, if applied, would preclude certain importers—particularly manufacturers—from operating efficiently. In these circumstances, investment laws and other incentive schemes under which the authorities are empowered to grant exemptions gradually become important policy instruments, while the customs tariff loses its importance. To a large extent, this has been the case in Senegal and the member countries of the Central African Customs and Economic Union (UDEAC).

These widespread exemptions and the low taxation of certain imports need to be re-examined in the light of the countries’ medium-term objectives. Thus, the exemption or low taxation of oil imports by some countries might have been justified before 1973 on grounds of the need to aid industrialization and the development of transport facilities, but it now runs counter to energy conservation efforts. In addition, it introduces distortions in the allocation of foreign exchange by making imported oil relatively cheaper than dutiable raw materials and intermediate inputs used by other industries. In some cases, it may even happen that preference is given to imported goods even when locally produced goods of equivalent quality are available, particularly if the exemption has resulted in removal of a protective duty. No less damaging is the zero rating of essential food. Although always predicated on equity grounds as a means of helping the needy, this policy encourages importation—and discourages domestic production—of staple foods such as rice, maize, and fish. Similarly, there is no good reason to exempt government imports, imports needed to carry out government contracts, or goods imported by quasi-government institutions. Such exemptions tend to decrease artificially the costs of government services and consequently to establish implicit subsidies, since the prices charged for those services will not reflect actual costs. It also encourages the use of imported materials and leads to abuses and leakages. The argument that this policy does not have a net revenue effect is, therefore, neither entirely correct nor relevant.

Contrary to the case of protection, where quick action is usually ruled out until a number of complex studies are completed, governments can generally take immediate steps to remedy, at least partially, the erosion of the taxable base caused by exemptions. It is not uncommon for governments in urgent need of additional revenue to move quickly to levy minimum or additional customs duties on oil and on essential foods. Such a move also helps a country to achieve its medium-term energy conservation objectives (in the case of oil) and to encourage its agricultural production (in the case of essential foods). Equally, exemptions on government imports and on imports needed to carry out government contracts can be removed without delay. In general, all exemptions should be reviewed with a view to removing them unless their raison d’être is beyond doubt. In doing so, however, international agreements, laws, and other legal arrangements that may have been used to grant exemptions over specified periods of time should not be violated. In these cases, it may be necessary to wait until such exemptions have lapsed.

Regional Groupings

Affiliation of many sub-Saharan African countries with regional agreements, such as customs unions and other arrangements, has inevitably limited their freedom to use customs tariffs as policy instruments. Examples of these arrangements are the Mano River Union Agreement, involving Liberia and Sierra Leone; the UDEAC, which includes Cameroon, the Central African Republic, the People’s Republic of the Congo, and Gabon; and the Economic Community of West African States (ECOWAS), which comprises Benin, The Gambia, Ghana, Guinea, Guinea-Bissau, Ivory Coast, Liberia, Mali, Mauritania, Niger, Nigeria, Senegal, Sierra Leone, Togo, and Upper Volta. Normally, these customs unions have common external tariffs, and tariff changes have to be effected through central committees on which all member states are represented. Agreement among members is not easily reached, and this has often led to inelasticity of customs revenue and has narrowed the base of import taxes.

Some of these regional arrangements contain provisions for the imposition of special taxes by individual member states, particularly where rate differentiation among them has seemed desirable. One example is the UDEAC’s complementary tax. Since these special taxes are the only instrument available to member countries for conducting a tariff policy of their own, they have been allowed to proliferate. This has resulted in a reduction in the uniformity of import taxation and in a distortion of the effects of the common tariff. Even when the regional agreements do not have these escape clauses, some member countries have introduced taxes on imports, disguised under a variety of names, to make up for the loss of revenue caused by adoption of the common tariff.

The advantages of regional integration are many and complex and cannot be evaluated solely from the viewpoint of taxation policy. Nevertheless, the existence of a customs union and a common external tariff should not be construed by the member states as an obligation to renounce sound principles of tariff reform. On the contrary, it seems imperative that the member countries pursue these principles relentlessly, working through available regional channels. Thus, simplification of the common tariff objectives, minimization of exemptions that are not sufficiently justified, and harmonization of tariffs with indirect internal taxes appear to be just as reasonable targets for regional groupings of countries as they are for a single country.

Import Taxes and Exchange Rate

In some cases, the authorities of a country may choose to raise import duties to compensate for an overvalued currency. Although the ultimate objective—to increase the cost of imports to a realistic level—is achieved without going through a politically less desirable currency devaluation, the action has a basic flaw; the tariff is not particularly well suited for this use. Not only does raising import taxes not help exports (in contrast to the effects of a devaluation), it actually raises their costs to the extent that exporters’ inputs are imported and that no refund or drawback system is available. However, the most important consideration is that manipulation of the tariff substantially distorts its protective function and may have other harmful allocative effects.

Administrative Considerations

Proliferation of Duties and Taxes

It is not uncommon in the sub-Saharan region to find that imports are subject to a variety of duties, fees, and other levies. For example, in Sierra Leone there is (a) a customs tariff, (b) two import fees (the invoice entry fee and the import licensing fee), and (c) a foreign exchange user charge levied on import licenses. In Somalia, imports are subject to an import duty (customs duty plus fiscal duty) and to an administrative and fiscal duty; in addition, there are wharfage and storage fees, and an ad valorem stamp duty. In UDEAC countries, a customs duty, an entry duty, a turnover tax on imports, and a complementary tax coexist.

This proliferation of import taxes is due mainly to ad hoc attempts to raise additional revenue by superimposing new charges over the original tariff, sometimes even using across-the-board flat rates. It is also the result of efforts to differentiate taxation of imports according to their origin in certain cases where there have been agreements with other countries to lower tariff barriers; the mutually agreed reductions would apply to some, but not all, of the various levies. A classic example is the experience of the UDEAC. Before the 1975 Lomé Convention, the UDEAC, as an associate member of the European Economic Community (EEC), exempted imports from the Community and its associated African countries from customs duties. For revenue purposes, the common external UDEAC tariff included another levy called the entry duty. Although the differentiation of imports according to origin no longer occurs, the two levies have continued to exist side by side. In some cases, charges such as statistical taxes, which had been originally introduced to defray the costs of certain services, became a part of the tariff’s rate structure and ceased to bear any relation to those costs. The rate structure of the tariff is often unnecessarily complicated, with negligible differences among rates. For example, there are 28 ad valorem rates in Sudan, including 6 that range from 5 per cent to 15 per cent. The same problem was observed in 1977 in Togo, where 9 rates of the fiscal import duty ranged from zero to 30 per cent.

This complex structure of import taxation greatly complicates administration and may result in inconsistent rate patterns. First, the calculations required to assess the tax liability are numerous, increasing the possibility of errors. Second, the complex structure makes it difficult to put together in one publication or volume the various taxes and duties on imports. The result is that self-assessment by importers or customs brokers—a labor-saving device for the customs department that becomes indispensable as the volume of trade grows—is not feasible. Third, proliferation of duties and taxes leads to extreme difficulty in readily identifying total tax burdens on particular items, thereby increasing the problems of rational policy formulation. It is also incompatible with the effort being made in most of the countries in the region to introduce a rational classification standard, such as the one provided by the Brussels Tariff Nomenclature.

To remedy this situation, the first step should be to consolidate into one or two charges the customs duties and all additional levies having an equivalent effect.5 It should then be possible to simplify the tariff structure by establishing a small number of even-modal rates and a maximum rate that would be in accordance with the rate distribution of the original tariff in the absence of new policy directives. In the long run, when—as discussed previously—the desired degree of protection to be given to each industry has been determined and other necessary changes have been introduced in the tax system, a comprehensive realignment of import duty rates should be undertaken in the framework of a comprehensive development strategy.

Use of Specific Rates

Generalized use of specific, rather than ad valorem rates, has also been an important factor—particularly in many nonfrancophone countries—contributing to the inelasticity of customs revenues. The case of Sierra Leone is an example. In the fiscal year 1977/78, as much as 15 per cent of this country’s total imports were subject to specific rates of customs duty. This preference for specific rates has usually been justified on grounds of administrative convenience, as these rates are applied irrespective of values and are therefore easier to administer than ad valorem rates. However, in times of rising prices, specific duties simply mean a steadily decreasing effective rate of taxation of imports, which tends to undermine the objectives of the original tariff, including that of protection. This is the reason that many countries around the world have increasingly resorted to ad valorem duties in the last 50 or 60 years.

The solution to the problem involves two separate steps. First, it is necessary to review carefully the justification for each existing specific rate. Then, when there is not sufficient reason to maintain a specific rate, it should be converted to an ad valorem rate; if, on the other hand, conversion to ad valorem rates is undesirable, a mechanism should be introduced to permit frequent increases in specific rates in order to maintain the level of import taxation in real terms as import values change.

In any case, even a limited minimum price list for valuation purposes (discussed below) is easier to update than specific duties and is, therefore, preferable to them.

Reliance on Administrative Values

Lack of adequately trained personnel has been the main factor precluding many sub-Saharan African countries from following sound principles of valuation. One of the most important functions of any customs administration is to make sure that imports are correctly valued in accordance with widely accepted methods. For this reason, standards such as the Brussels Definition of Value and, more recently, the General Agreement on Tariffs and Trade (GATT) Definition of Value have been developed by international agencies. Unfortunately, application of these standards requires a high degree of technical knowledge and experience on the part of customs personnel. This requirement has proved extremely difficult to fulfill.

To prevent underinvoicing, many countries have therefore resorted to establishing price lists, against which the values declared by importers or shown on the respective invoice are checked.6 If the listed price is greater than the value obtained from the importer, the former is used for customs purposes. Such systems are simple to administer, but they have shortcomings. One relates to the degree of trade unfairness that these systems make possible. If a country that uses a list of administrative values decides to confer a high level of protection on a certain item that is domestically produced, all it has to do is inflate artificially the list price of competing imported products. This obviates the need to go through the highly visible process of raising the tariff level. However, the most serious shortcoming, particularly from the stand-point of revenue, is that the list has to be updated at frequent intervals, especially at times of worldwide inflation. If this updating is not done, revenue erosion occurs in much the same way as when specific rates of duty are not adjusted for price changes. The cases of Somalia and Chad illustrate this. The list used in Somalia in March 1980 had not been updated since February 1977. Meanwhile, wholesale or industrial prices in the countries from which Somalia’s imports mainly originate went up by between 9 per cent (Federal Republic of Germany) and 39 per cent (Italy) over the period 1977–79. Equally, in Chad, standard values established in 1970 were still being used in late 1976.

In the long run, all customs administrations will have to rely on internationally accepted valuation standards if they are to avoid disputes and minimize differences of opinion with trading partners. Because it will take time to familiarize the trading community with the new standards and to establish the administrative capability required to apply them, it would be wise to start as soon as possible. Recruitment and training of qualified personnel, introduction of needed legal changes, setting up of a valuation unit within the customs department, and compilation of information that is relevant for valuation purposes are all tasks that cannot be performed over a short period. In view of this, it is generally admitted that some countries will have to continue using minimum price lists in the near future. In these countries, use should be restricted to goods known to be under-invoiced, and every effort should be made to set realistic prices. To avoid the revenue-eroding effect mentioned above, however, it is imperative that the price lists be updated frequently. This could be done by compiling comprehensive and reliable data on prices from foreign trade journals, retail and wholesale catalogs, and other sources. In addition, customs officers should be aware of the value of their own daily work for purposes of updating the price list.

Customs Procedures and Personnel Problems

Many countries of sub-Saharan Africa have paid attention to the organization of customs departments and to customs legislation in general. It is in the areas of customs procedures and personnel matters where the most serious deficiencies are still discovered. A comprehensive review of these aspects could suggest changes that would have a meaningful positive effect on the efficiency of the customs operations.

Included in this review should be adequacy of warehousing provisions and charges for allowing surveillance of goods, as well as availability of bonded warehousing; controls for duty-free concessions, temporary importation, and transit; prevention of leakages through provisional clearance under guarantee, and refunds and drawback schemes; quality of the examination of goods for clearance; quality of the systems for payment of duties; arrangements governing the activities of customs agents; methods used for lodgment, registration, and postaudit of import entries’, flow of documents within the customs administration; and efficiency of the statistical reporting system. Once the review of these and other procedures has been concluded and shortcomings have been identified, new procedures should be designed and implemented. Of paramount importance in avoiding failure is the introduction of training programs. These should have detailed instruction manuals, prepared with a view to ensuring uniform application of the new procedures throughout each country. Other procedural changes will be necessary upon reform of the system of customs valuation along the lines discussed above. Finally, much more attention should be paid to increasing the salaries of the staff and improving their working conditions. Failure to do so will only result in increased corruption—to which customs work is particularly vulnerable—or in an acceleration of the exodus of the best-qualified and best-trained personnel to the private sector. In the final analysis, the success or failure of carefully planned policies depends to a large extent on the suitability and motivation of the staff responsible for carrying them out.

In 15 countries in the region import duties accounted for more than 40 per cent of total tax revenue, and in 24 countries the proportion exceeded 30 per cent (see Part II).

The term “customs duties” in this study is the equivalent of import duties.

Quantitative restrictions and exchange controls have also been widely used as protective devices.

A relatively simple measure of the effective protection rate for a particular product is given by a fraction whose numerator is the difference between the tariff rate on the product and the average tariff rate on the inputs used in its production weighted by the latter’s share in the product price on the world market and whose denominator is the percentage of value added per unit of output measured in world market prices.

In some cases, a second duty is necessary to allow for agreements already negotiated with other countries or to give the country an advantageous point of departure for future negotiations of tariff reductions.

In countries with exchange controls, insofar as for many relatively low-duty items the foreign exchange premium may be higher than the import tax penalty, overinvoicing may be widespread. Since this deviation from actual prices is only in the direction of enhancing customs revenue, it is not discussed here.

    Other Resources Citing This Publication