HIST363: Global Perspectives on Industrialization

Protection policy is frequently advocated as a means to promote industrial development. It can certainly do this for import competing activities, such as production of consumer goods. However, this production pattern locks producers into small, and typically slow growing, markets for their output. Further, it introduces a major discontinuity. Under a protectionist policy regime, an exporter must not only have sufficient comparative advantage to be able to compete in world markets, but must have sufficient advantage to be able to compete at world prices despite the cost increases resulting from protection levied on its intermediate inputs, and the adverse effects of real exchange rate devaluation on its costs for factors and nontraded goods.

Developing countries have increasingly come to recognize the adverse impacts of protection on their export performance. and begun to adjust their policies towards more open trade regimes. The most profound and far-reaching manifestation of developing countries' interest in greater participation in trade is evident from the wave of unilateral trade reforms that has swept the developing countries. These reforms have affected all regions, and all of the major types of policy distortions.

As discussed in Global Economic Prospects 2001 and presented in Figure 4, average tariff rates in developing countries have halved, from around 30 percent in the early 1980s, to around 15 percent in the late 1990s. By 1999, Francis Ng estimates that the simple average tariff in developing countries has fallen to 11 percent from The absolute reductions in tariff rates in developing countries have been much higher than in industrial countries and, of course, decreases from a higher level are likely to have a much greater welfare benefit than corresponding decreases from a lower base. In addition, the dispersion of tariff rates, which typically increases the welfare cost of any given average tariff rate, was substantially reduced.

Figure 4. Tariff rates since the early 1980s

One must be careful when examining changes in tariff rates, because a decline in tariffs may reflect substitution of nontariff barriers for tariffs. However, during this period, the coverage of nontariff barriers, including state trading monopolies, in developing countries also appears to have fallen considerably, as is evident in Table 2.

Table 2. Frequency of total core nontariff measures in developing countries, 1989–98

Note: Figures in parentheses are the number of countries in each region for which data are available.

Another important dimension of reform has been a sharp reduction in the number of countries using foreign exchange restrictions on current account, and in the average foreign exchange premia. The World Bank reports that the number of developing countries applying foreign exchange restrictions on current account has fallen sharply.

Table 3 shows foreign exchange premia for a range of countries in the 1980s and 1990s. This table highlights two things. Firstly, that average foreign exchange market distortions were enormous in the 1980s, and that these premia in most developing countries, in most regions, have fallen to very low levels. While the simple average foreign exchange rate premium is highest in the Middle East and North Africa, at 46.5 percent, this high rate is almost entirely due to large premia in Algeria and Iran. If these two outliers are excluded, the average rate falls to only 1.4 percent. When Nigeria is excluded, the average premium in Sub-Saharan Africa is less than 10 percent, down from 112 percent in the mid 1980s. Clearly, for most countries, the premia are now small enough to imply that foreign exchange distortions impose relatively small taxes on trade.

Table 3 Average black market premium (percent)

Notes: ^aSample of 41 developing countries. ^bAlgeria and Iran

There are good reasons to expect that, in this situation, a high protection regime will lock countries into continuing dependence on resource-based commodities which are typically less dependent on purchased intermediate inputs than is manufacturing, particularly in this era of production fragmentation. To allow further examination of this difference, Table 4 presents for a number of countries data on the cost structure of output and the effective rates of protection imposed on export-oriented activities. A striking feature is the top section of the tables is the much lower dependence of primary agriculture and resources commodities on intermediate inputs. This gives resource-based activities an opportunity to survive even in situations of very high protection.

Table 4. Shares of intermediate inputs in total costs, 1997

It is, of course, possible that the greater vulnerability of manufacturers and agricultural processors to high protection regimes would be offset by a type of tariff escalation that involves lower than average protection on intermediate inputs to agricultural and manufacturing sectors. To see whether this is the case, the second panel of Table 5 examines the effective rates of protection applying to exporters. These effective rate calculations are done very simply, taking into account only the effects of intermediate input shares and tariff rates.

They therefore ignore the additional burdens imposed on exporters by nontariff barriers on inputs, or by the real exchange rate appreciation associated with protection. What the results of these calculations strongly suggest is that the pattern of tariff protection does not provide any relief to exporters of manufactures or processed agricultural products. In fact, it appears that the pattern of real-world protection adds to the discrimination against exporters of manufactures and processed agricultural products resulting from their greater dependence on intermediate inputs.

Table 5. Effective Rates of Protection for Import Competing and Exporting Activities, 1997

Where the negative effective rates of protection seen in the lower panel of Table 4 the structure of protection is clearly a daunting problem for putative exports of manufactures or processed agricultural products, particularly if there are fixed costs involved in entering export markets. However, it is clear that this problem is much more manageable in many countries than it was in the early 1980s. If we triple China's protection level from its 1997 base to align it with the tariff rates that applied in China in the early 1990s, we find negative ERP-X's of –78 percent for processed food and –62 percent for labor-intensive manufactures. And this is before the direct adverse impacts of licensing, quotas and nominal exchange rate overvaluation, and the indirect effect of real exchange rate appreciation, all of which applied at that time, are factored in.

Another issue to be considered when evaluating the effects of an import-substitution regime on industrial structure and outputs is the incentives created for import-substituting activities. These results make it clear that the protection regimes prevailing in the selected countries in 1997 provided strong incentives for undertaking import-substituting activities despite the very large reductions in tariffs that had taken place since the 1980s.

The greatest incentive, on average, was provided in agricultural processing, where a high intermediate input share makes effective rates high on average, and particularly variable. Agriculture, and both manufacturing sectors also have high and variable effective rates for import-competing activities.

The large gaps between the effective rates of protection for import-competing activities and export-oriented activities create particular problems for attempts to develop competitive industries behind tariff walls, and have them develop into export-oriented activities. Frequently, the subsectors receiving high protection for import-competing activities will not be those in which the country has a comparative advantage. Even if the country does have such a comparative advantage, there is no guarantee that the protected sector will be successful in exporting over the anti-export barriers created by protection on its inputs and real exchange rate protection.

This sharp discontinuity in the incentive Direct evidence on the implications of increased openness for exports of manufactures is provided by a wide range of empirical studies using traditional CGE models. A notable study by Whalley pointed out that nondiscriminatory liberalization by developing countries could be expected to increase both manufacturing exports in total, but also south-south trade in manufactures. More recent studies of trade liberalization have pointed to large increases in developing country exports following liberalization, even from the much lower tariff levels prevailing after the Uruguay Round.

Studies using other methodologies have also highlighted the consequences of trade reform. A recent econometric study by Elbadawi, Mengistae, and Zeufack builds on recent economic geography models developed by Redding and Venables and concludes that increasing openness in African countries would considerably expand exports of manufactures.

Overall, it seems highly likely that the sharp reductions in developing country trade distortions since the early 1980s have played a vital role in allowing developing countries to so sharply increase their exports of manufactures, and hence reduce their dependence on resource-based products.