Chapter

4 Trade Regimes and Export Strategies with Reference to South Asia

Editor(s):
Paul Streeten
Published Date:
September 1988
Share
  • ShareShare
Show Summary Details

I. Introduction

The issue of structural adjustment and growth in developing countries essentially involves changes in domestic production possibilities relative to trade opportunities. This paper is concerned mainly with which sectors a country ought to encourage, whether there should be recourse to foreign trade, and what policy instruments should be chosen. There are two strands in the literature dealing with these issues. Trade theory has traditionally been concerned with the choice between domestic production and exports or imports. And the central questions in the public economics literature concern what to produce, what instruments to choose, and the consequences of policy for households and government revenue. There is thus an overlap between the two approaches, and Subsection II. 1 juxtaposes the main findings of each with respect to the selection of trade regime—including the arguments for free trade and the case for protection—and the choice of instruments.

Policymakers and economists in developing countries have been much concerned with building up “appropriate” domestic productive capability to ensure growth (defined in terms of gross national product (GNP), GNP per capita, living standards, and so on). And the “development literature” has also addressed the problem of trade versus domestic production—given the vagaries of the international economic environment—and problems of adjustment to movements in the terms of trade. Subsection II.2 refers to some of the debates in this strand of literature, including the discussion of the possibility of export-led growth.

Trade policy in the Indian subcontinent during this century has illustrated many of the policy options debated, from laissez faire to autarky, as well as some of the problems with policy instruments, such as quotas and import licensing. Section III deals with the choice of tariffs or quotas for protection and possibilities for export promotion with reference to the experience of countries in South Asia, with illustrations mainly from India, but also from Pakistan and Bangladesh. It is not intended to provide a comprehensive survey of the empirical literature relating to these three countries.

Section IV draws on work done on India in the late 1970s that uses a system of economy-wide shadow prices to show which sectors might be encouraged to permit adjustment with growth in output, given a concern for the welfare of the poor. Section V contains some concluding remarks.

II. Growth, Trade, and Theory

The relationship between growth and trade has been the subject of some controversy in the development literature, and this has been extensively surveyed in recent papers by Bliss (forthcoming), Krueger (1984), and Findlay (1984). Section II. 1 contains a brief review of the case for free trade and arguments for restrictions from the theoretical literature of international trade and public economics. The gains-from-trade proposition is put in the context of developing countries. Section II.2 examines arguments put forth in the development literature for restricting trade, including early import-substitution theories; export pessimism, which has been a recurrent theme; and infant-industry arguments. A synthesis of the trade and development literatures suggests that the gains from trade need not be synonymous with free trade and that, in general, some trade is better than no trade.

1. The Gains from Trade

The central proposition from normative trade theory is that there are gains from trade and that free trade is Pareto superior to autarky under certain assumptions and is also superior to various forms of trade restriction. (The developments in the normative trade literature are reviewed in Corden (1984).)

The gains-from-trade proposition rests on very simple arguments. Under autarky, a country’s consumption possibilities are limited by its production-possibility frontier. These consumption possibilities are increased by net trade with the rest of the world. Thus, with trade, the consumption-possibility frontier lies outside the country’s production-possibility frontier and touches the latter only where the marginal rates of transformation (MRT) in domestic production equal those attainable with trade. For the small-country case, the marginal rates of transformation are given by world prices. The gains-from-trade result could be extended to the large-country case—where a country has monopoly power in trade and the marginal rates of transformation no longer equal the price ratios— by applying the optimal tariff at the chosen point. Consequently, feasible consumption in autarky can be dominated by combinations of trade and domestic production.

Although the early proofs for the potential gains from trade were based on a number of restrictive assumptions—absence of increasing returns, no distorting domestic taxes, no externalities, the feasibility of lump-sum transfers, and flexible factor prices that ensure full employment of factors—recent work has relaxed some of these assumptions. Dixit and Norman (1980) show that lump-sum transfers are not necessary for the gains-from-trade result and that factor taxes, including the income tax, will suffice. Ohyama (1972) examined the gains from trade in the presence of tariffs and subsidies, and the condition for a welfare improvement from a move to such a distorted situation is that net revenue be positive. Thus, trade under self-financing subsidies is preferable to autarky. The introduction of uncertainty has been shown by Helpman and Razin (1978) to be like an adverse movement in the terms of trade, although gains from trade exist. And in the intertemporal case, it has been shown by Smith (1979) that gains from trade exist, as in the static case, subject to similar qualifications: that balanced trade in goods at each date is preferable to autarky and, further, that when international borrowing and lending are permissible, a small country will gain further if trade is balanced over time in terms of present discounted values.

The literature concerning trade and imperfect competition is surveyed in Dixit (1984). For a pure domestic monopolist or price leader in an import-competing industry, trade serves to limit monopoly power, and protection is harmful. Where an exporting firm has monopoly power in the home market, it is likely to have less monopoly power in export markets. While the firm would wish to charge higher prices at home than abroad, the home country’s interests are in having marginal-cost pricing at home and monopoly pricing in the export market. When a foreign firm has monopoly power in the home market, protection provided either by imposing import tariffs or increasing domestic firms’ profits would appropriate some of the pure profits earned by the foreign firm, although loss of the home-country consumer surplus would have to be considered also. The case of cartels and the producing country’s export tax policy are considered by Dixit and Stern (1982). Monopolistic competition has also been considered in the literature. It makes a distinction between intra-industry trade (based on product diversity and scale economies) and interindustry trade (based on factor endowments). The result that intra-industry trade should occur in countries that are similar, leading to a greater product variety, is what has been observed empirically. Dixit (1984) shows that policies under oligopolistic conditions vary from those under competitive conditions and that trade restrictions might be desirable, since prices are set above marginal cost. However, this is an area of active research, and it is still to be established whether other policies might achieve the same result more efficiently. Moreover,

vested interests want protection, and relaxation of anti-trust activity, for their own selfish reasons. They will be eager to seize upon any theoretical arguments that advance such policies in the general interest. Distortion and misuse of the arguments is likely, and may result in the emergence of policies that cause aggregate welfare loss while providing private gains to powerful special groups.1

Newbery and Stiglitz (1981, 1984) compare autarky and free trade between two competitive, but risky economies with no insurance markets and show that free trade may be inferior to no trade. However, some trade is shown to be preferable to no trade. The policy choice of trade restrictions, tariffs, or quotas is examined in Section III.

The case of “immiserizing” growth was brought back into the discussion by Bhagwati (1958). (It had been considered, in a less general setting, in 1894 by Edgeworth, who referred to it as “damnifying.”) Thus, national welfare declines with growth in national income, provided that the decline in the terms of trade exceeds the favorable effects of the expansion at constant relative product prices. However, a country that could affect its terms of trade would levy an optimal tariff, and immiserizing growth would not occur.

Another type of “immiserizing” growth was considered by Johnson (1967). With fixed terms of trade, in a two-sector, two-factor open economy, a tariff or trade distortion that results in the output of the import-competing good being too large, and the output of the exportable good being too small, may lead to the immiserization. This occurs because factor endowments change so as to expand the inefficient protected sector at the expense of the efficient export sector. This result also applies to the trade in factors. In the model of Brecher and Bhagwati (1981), where some of the factors of production in the home country are owned by foreigners, the optimum tariff vector must be worked out simultaneously for goods trade and foreign factor income, since otherwise immiserizing trade can result. See Bhagwati and Srinivasan (1983), chapters 16–25, for a discussion of growth, comparative statics, and many kinds of distortions.

Dixit (1985) surveyed the modern public finance approach to the open economy. International trade—which represents a new set of transactions, possible externalities, and distortions—enlarges the consumption-possibility set. Government objectives may still be characterized as raising revenues, increasing household welfare, and facilitating production; and there may also be other, non-economic objectives or constraints. The policy instruments are the set of taxes and subsidies on domestic or foreign transactions and activity, subject to administrative feasibility. And central features of the public finance approach are that tariffs affect domestic-resource allocations and income distribution and that domestic taxes affect trade, given the interdependence of the economic system. (See also Stephen Lewis (1984).

A major result from the public finance literature has a bearing on trade policy. This is the application of the Diamond-Mirrlees (1971) aggregate-production-efficiency theorem, which implies that marginal rates of transformation between domestic production activities and foreign trade should be equalized. In the case of foreign trade, these equal world prices, except in those cases in which there is monopoly power, where the optimum tariff will yield the necessary equality. For domestic production, marginal rates of transformation equal producer prices. A consequence of the efficiency theorem is that there should be no taxation of (or subsidy on) producer or intermediate goods. This paper will return to this result in discussing tariffs and protection in Section III.

There are two results from international trade theory that carry over to the public finance analysis of trade. First is the Bhagwati-Johnson principle of targeting, according to which, if there is a distortion, it should be countered directly; or if a distortion is to be introduced in an effort to achieve a non-economic objective, then a tax instrument should be used that acts directly on the relevant margin. And if there is an external economy in production, the first-best policy is to use an appropriate Pigovian subsidy. “It is only if this is impossible that the indirect effect of a tariff to stimulate domestic production can be useful as a second-best (or worse) policy.”2 This has a direct bearing on the choice of instruments once it can be established that a particular industry or sector should be encouraged domestically. The second result is that if particular groups are affected adversely through trade, then domestic goods or factor taxes or subsidies, rather than tariffs, should be used to compensate them.

The gains from trade have been established in principle. However, these are only potential and may not be realizable per se. It is possible, however, for laissez faire to lead to Pareto-inefficient equilibria in the face of market distortions or policy failures. In such cases, adding another distortion or restriction, such as prohibition or taxation of trade, may prove beneficial in accordance with the general theory of the second best.

2. Development and Trade

Although the burden of the theoretical literature surveyed above suggests that some trade is better than no trade, and that free trade might Pareto dominate both under given circumstances, development economists have remained divided over the issue of trade, and there is considerable skepticism concerning feee trade. In this subsection, two sets of arguments used for justifying departures from free trade are examined: these may be broadly classified as the “export-pessimism” and “infant-industry” cases. Shadow prices, which provide a guide as to which industries should be encouraged, are also discussed briefly.

a. Export Pessimism

Pessimism concerning developing country exports to developed countries underlies some of the demands for import substitution that have been made since the 1950s. At the extreme there is a claim, associated with Prebisch and Singer, that the terms of trade must inevitably move against primary products in favor of manufactures. And Nurkse, in considering the balance between the global demand and supply for primary products, concluded that developing countries generally cannot expand primary product production without suffering terms-of-trade losses. This is because a “rational” strategy by a poor, small country to expand production of the (primary) good(s) in which it has a comparative advantage might lead to global overproduction of such good(s) or to immiserization of a group of such countries. While a solution to the problem of overproduction of primary commodities may lie in collaborative behavior, the problems encountered in reaching agreement on policy recently faced by the Organization of Petroleum Exporting Countries (OPEC) (which has been the most successful such experiment) do not bode well for other commodity-producing cartels.

Developing countries are often characterized as primary producing countries of the South that are dependent on their trade with an industrialized North. Terms of trade “play a key role . . . as the regulator that makes the growth rate of the South conform to the exogenously given long run growth rate of the North. Trade is the ‘engine of growth’ for the South, but the pace of the engine is set by the growth rate of the North.”3 A number of North-South models are reviewed in Findlay (1984), as are variants of “unequal-exchange” models that may be seen as special cases of such North-South models. However, the proposition that there has been a secular decline in the terms of trade for developing countries has been challenged by Krueger (1984), among others. This is because it is difficult to establish which commodities are exported by poor countries and which by richer countries; and, indeed, a classification of countries as poor and rich is problematical. Also distinguishing between cyclical and secular trends is particularly difficult, since the results obtained are strongly affected by the choice of initial and terminal periods, and the data are often of poor quality. Krueger ((1984), p. 560) discusses unpublished work by Michaely which showed that for the years 1952, 1955, 1960, 1965, 1970, and 1973,

the price of exports of the poor countries rose more than that of the rich; the price of imports of the rich countries rose more than that of the price of the poor; and the ratio of the two moved in favor of the poor countries. Hence, the terms of trade had necessarily moved in favor of the poor countries over the period covered by his data. He then proceeded to show that using the conventional measure of manufactured goods prices relative to primary commodity prices provided the “orthodox” Prebisch-Singer result.

No doubt critics could apply the Krueger strictures to the Michaely analysis and argue about choice of period, countries, and commodities. However, the point remains that it is simplistic to equate developing countries with primary producers and vice versa.

In his Nobel lecture, W. Arthur Lewis (1980) also argued that over the past century, the “rate of growth of output in the developing world has depended on the rate of growth of output in the developed world”4 and that the principal link between the two growth rates is trade. On the assumption that growth rates in developed countries would be lower in the future than in the immediate postwar era, and that the prospects for developing country manufacturing exports would be further limited by Organization for Economic Cooperation and Development (OECD) tariff and nontariff barriers, the future of developing country growth rates would depend on those diverse countries, such as India, that have the capacity for self-sustaining growth and trade with other developing countries. Thus growth rates in leading developing countries would substitute for growth in the developed countries and provide a demand for the exports of other developing countries.

The empirical evidence examined by Reidel (1984) suggested that many developing countries (and particularly those in East Asia) diversified into manufacturing exports to developed countries and that only a few countries, mainly in Africa, were exporting a single primary product. The share of manufacturing exports (in total developing countries exports) had increased from 7 percent in the mid-1950s to over 20 percent by the late 1970s, and the proportion of developing country manufactured exports going to the developed countries increased from 45 percent in 1955 to over 60 percent in 1978. However, it should be borne in mind that 60 percent of developing country manufactures exported in 1978 were from the Republic of Korea, Taiwan Province of China, Hong Kong, or Singapore. And as Fields (1983, 1984) points out, these countries are characterized by competition in the labor market, which clears through wage adjustment. Thus, relatively low unit-wage costs (in terms of levels of productivity and the nominal exchange rate) and an efficient functioning of the labor market facilitate an export-led growth strategy. Fields contends that the Jamaicas, Mexicos, and Indonesias of the Third World have wages in the exporting sectors that are two or three times the market-clearing levels and that these “countries start out at an enormous disadvantage in trying to compete successfully in world markets with the U.S., the European Economic Community, Japan, and the East Asian NICs [newly industrializing countries]. . . . Not to be able to export profitably is bad. To export unprofitably is worse.”5 Thus, the overall picture indicates that for individual countries there are possibilities of exploiting trading opportunities, and that there may be scope for extending exports to other developing countries if OECD policies are too restrictive. However, this conclusion assumes “appropriate” domestic policies and does not suggest that developing countries export solely for the sake of exporting.

b. The Infant-Industry Argument

One of the oldest arguments for an exception to the free-trade proposition is the case for infant-industry protection. Corden (1984) equates this with an argument for temporary protection to account for some market imperfection or externality. A tariff to correct for the distortion, in comparison with a production subsidy, would create a consumption distortion, and a production subsidy would be preferable if only these two instruments were available. However, more direct methods might be preferable to the production subsidy, as the following examples indicate.

Where there are dynamic internal economies, as in a firm or an industry going through a period of investment in human capital, the learning benefits stay within the firm or industry, and it may not be possible to finance investment in such an industry, given capital-market distortions. The first-best policy is to improve the capital market and then subsidize the factor or input that gave rise to the internal economy. A general output subsidy may fail to correct the capital-market distortion or to encourage the dynamic factor sufficiently. A tariff would create further distortions, and an export subsidy would be the least preferred policy option.

In the case of dynamic external economies, given mobility and market imperfections, such as capital-market imperfections or wage rigidities, the effects of labor training would not be internalized by the firms at the infant stage or any other, since trained workers would move to better-paid jobs quickly. The preferable policy options would be to improve the capital markets, and to finance or subsidize the labor training. Less attractive would be the subsidizing of the employment of labor in those sectors that provided more trained manpower than others. Least favorable would be an output subsidy.

A third argument that has been discussed extensively concerns knowledge diffusion, also an example of a dynamic external economy. This diffusion is rapid in the infancy stage, although it is in the private interests of firms to restrict the diffusion.

There is little direct evidence about the behavior of specific infant industries in developing countries: estimates of the pattern of cost reductions and benefits over time and the duration of infancy. Krueger and Tuncer (1982), however, found that in the Turkish case, more protected industries did not have greater reductions in costs than less protected industries. And there is a strong presumption that because high levels of protection have continued in developing countries for long periods, “protection in developing countries generally has not been justified on infant-industry grounds” (Krueger (1984), p. 525). Despite the presumption that the period of infancy should not exceed five to eight years (Balassa (1975)), fragmentary historical evidence suggests that the Japanese cotton textile industry took two to three decades to mature, the Japanese automobile industry three to six decades, and the Korean textile industry about four decades. However, the period of maturation for some Korean industries has been as short as a couple of years, as in the shipbuilding industry. And the rapid industrialization and diversified exports of countries such as the Republic of Korea and Brazil show that the period of maturity can come about reasonably quickly, although the differences between the Korean textile and shipbuilding industries tend to suggest that there may be significant differences across sectors. (See Bell, Ross-Larson, and Westphal (1984) for further references.)

c. Shadow Prices

The question of which industries or sectors ought to be encouraged involves issues of the intertemporal social costs and benefits of the proposed policy change. This essentially involves the use of shadow prices, defined as the increase in social welfare resulting when an extra unit of public supplies becomes available. There has been a voluminous literature on the concept of shadow prices, and a recent integration of the theory has been provided by Dreze and Stern (1985). A natural application is to the theory of reform in which the planner inherits an environment—including distortions, warts, and all—and pursues policy objectives within a given area of control. (For a statement of the theory of reform in a simple context, see Ahmad and Stern (1986 b) and Drèze and Stern (1985); for an application, see Ahmad, Coady, and Stern (1986)).

Although shadow prices have been calculated for many economies, since they are extensively used in project appraisal, care should be taken to ensure that these are consistent with the models used for the reform discussion, for “when the social value of projects depends upon the ‘reforms’ accompanying them, projects can no longer be evaluated in isolation.”6

The shadow prices essentially capture the full general-equilibrium effects of a policy change on welfare. And, in principle, they should be derived from a fully articulated general-equilibrium model of the economy. However, the shadow prices provide summary statistics for policy from the full model and are often more flexible and more easily understood than the full model. Thus, consistent with a plausible set of shadow prices, it should be possible to construct a general-equilibrium model and to make appropriate welfare judgments. For example, if certain policies are known to affect the shadow wage, then one can examine the consequences for trade and tax policy fairly quickly, although it may prove difficult to modify the general-equilibrium model to incorporate the changed circumstances.

The use of world prices as shadow prices goes back to the Little-Mirrlees 1969 OECD Manual (revised version provided in Little and Mirrlees (1974)) and has been the subject of much discussion. (See Bliss (forthcoming) and Corden (1984) for reviews of the recent literature.) The use of marginal border prices as shadow prices for tradables is fairly robust and extends to the case where world prices are affected by domestic policy, given optimum tariffs. And the presence of nontradables does not affect the rules for shadow prices for tradables. (See Dixit (1984).) Price wedges, such as tariffs, do not affect the classification of tradables, although quantitative restrictions, such as binding quotas, imply that at the margin, a commodity is not traded and should be treated correspondingly. In the absence of distortions, the shadow prices of nontradables would equal their domestic market prices. With distortions, the two can differ and may be negative, as shown by Bhagwati, Srinivasan, and Wan (1978). The Ahmad, Coady, and Stern (1984, 1986) estimates of shadow prices for Pakistan and India use world prices for tradables and take into account the distorting effects of trade restrictions on prices that producers face, imperfections in factor markets, and premiums on savings. Different sets of shadow prices were calculated, using input-output techniques in large part, that corresponded to different assumptions governing the classification of goods as traded and nontraded and to different valuations of factors. Some of the results are discussed below.

III. Tariffs, Quotas, and Trade Policy in South Asia

The laissez-faire approach of the British authorities to economic policy during the colonial period has been blamed for turning India into a market for British manufactures and a source of supply of cheap raw materials. “For many years, Indian tariffs were kept low, and the overall tariff structure afforded minimal protection for Indian industries, thus strengthening the complementary position of the Britsh and the Indian economies.”7 However, the case for tariff protection was conceded after the First World War. The ensuing pattern of industrialization is documented in Bhagwati and Desai (1970). Direct import controls, however, were introduced during the Second World War to regulate foreign exchange and shipping space for the war effort and for essential civilian supplies. These controls continued in various forms until India achieved independence, and these have been a feature of policymaking in both India and Pakistan since. (For India, see the Ministry of Finance’s Report of the Committee on Controls and Subsidies (or Dagli Committee) (1979)) and Lal (1980); and for Pakistan, for the period up to 1970, see Islam (1981).)

The experience of wartime controls, as well as colonial laissez-faire, led Indian policymakers to adopt an import-substitution strategy after independence that allowed for imports of raw materials and intermediate goods while tightening control of other imports. With the analytical planning framework of the Mahalanobis model, the Second Five-Year Plan continued the import-substitution policies. The imports of capital goods and raw materials increased until they were checked by a foreign exchange crisis in 1958, which led to a drastic reduction in imports through extensive controls. A tight import policy continued into the mid-1970s.

With the pursuit of the import-substitution policy, it soon became apparent that an overvalued exchange rate and import controls discriminated against exports. Thus, explicit policies for export promotion were introduced in the early 1960s, although these were related to the system of import controls. The extant export subsidies in the late 1970s were described by the Dagli Committee as having been based on “‘hit or miss’ methods, with an eye to securing a 7 to 8 per cent rate of growth of overall exports, without reference to either the costs or the long term benefits.”8 More recently, the Indian Ministry of Commerce’s Report of the Committee on Trade Policies (Hussain Committee) (1984) has argued that while a country like India cannot have export-led growth, there should be a “quantum jump” in exports, and that the emphasis of import policy “should move from import substitution per se to efficient import substitution.” While agreeing that Indian import substitution in some sectors has been too costly and that there is a need to improve export performance, Chakravarty (1984) expressed reservations about export-led growth, on the East Asian pattern, in the Indian context. This was because of (i) the relative size of the nontraded goods sector; (ii) the limit to which efficiency wages could be reduced to ensure competitiveness; and (iii) the uncertainty concerning the changing international division of labor (an example of the export pessimism that was discussed in Section II) meant that the prospects for greatly increasing exports were limited.

With the switch to an import-substitution policy after independence, the Indian Government, for example, had a limited number of policy instruments in hand with which to implement these policies. In this connection, both tariffs and quotas have been used, along with investment licensing and price and output controls. Subsequently, measures to counterbalance the bias against exports arising from the instruments used to encourage import substitution came into being in the early 1960s. The standard equivalence theorems in the trade literature concerning tariffs and quotas are discussed in Subsection III.1.

While tariffs are an instrument of commercial policy that guide the pattern of domestic investment allocation, it is known (Tanzi (1987)) that customs duties are convenient “tax handles” and that revenues from international trade are easier to collect than general sales taxes or taxes on income. Thus, in India, customs duties averaged more than 40 percent of the tax revenue of the central and state governments until the mid-1950s. However, with the growth of the domestic production base, and the introduction of quotas and direct controls on trade, customs duties had declined in importance relative to excises on domestic production by the early 1960s (with the latter providing 45 percent of total tax revenue in 1959/60, as compared with 20 percent from customs duties). By the 1980s, customs still accounted for around 20 percent of total tax revenue—a not unsubstantial amount of revenue, given that tax revenues in India by the late 1970s were of the magnitude of 19 percent of gross domestic product (GDP). (This proportion had also been projected in the 1986/87 budget estimates. See the Indian Ministry of Finance’s Economic Survey, 1985/86.) Since revenues are of obvious importance to countries like India and Pakistan, the prevalence of quantitative restrictions and quotas in the subcontinent needs to be examined in this context. Moreover, the major tax tools, customs duties, and excises in both countries bear quite heavily on intermediate goods and raw materials, causing problems of possibly unintentional taxation of final goods and exports.

After brief discussions of the trade control instruments, tariffs, and quotas in Subsection III. 1, and of protection in Subsection III. 2, the paper proceeds to an assessment of export policy in India in Subsection III. 3.

1. Tariffs and Quotas

A decision by, say, the Bangladesh Government to protect the domestic steel industry—for example, the Chittagong Steel Mill— either by tariffs or quotas on imports, has an immediate impact on domestic users of steel through its increased price, and if some of these firms are exporting establishments, then the import tariff on steel is like an export tax. In this section, some of the important “equivalence results” from the trade literature are discussed.

Of policy importance is Lerner’s symmetry proposition, which in a two-good real model postulates that since both an import tariff and export tax lower the domestic price of the exportable relative to the importable, they have the same impact on production and consumption. This result has been extended to the case of several goods, where relative prices matter, and the treatment of nontradables introduces particular exchange rates. (See, for example, Corden (1984) or Stephen Lewis (1984), Part IV.) This equivalence result has important implications for countries that rely heavily on import duties for revenues, since these duties have the same effect as export duties.

Within the Walrasian general-equilibrium context, each quota or nontariff barrier has its shadow price, which could be replaced by an equivalent tariff and, abstracting from income effects, if this were done, the equilibrium would not be affected. The main difference, even in this context, would be in possible differential distributional effects if quota rents were to go to nongovernment agents, whereas it is presumed that tariff revenue accrues to the government. Thus, in cases where the quotas were not auctioned domestically, there would be a revenue loss to the government. Further if the quotas are allocated inefficiently by the bureaucracy, and if there is no resale, further deadweight losses may occur that would have been avoided by the imposition of a tariff. There is evidence that in both India and Pakistan, there has been a differential access to quotas and that the inefficiency referred to above is quite common. Thus, even in the general-equilibrium sense, a tariff may be preferable to quotas or quantitative restrictions.

Non-equivalences between tariffs and quotas in the case of monopoly have been extensively discussed in the literature, following Bhagwati (1965). This occurs when there is either actual or potential monopoly on the part of domestic import-license holders, domestic import-competing producers, or foreign suppliers. Thus, a quota that has the same effect as a tariff on the amount of imports may differ with respect to the effects on domestic prices and production. A case in point is when a quota, while raising the domestic price, provides a degree of monopoly power to the domestic producer, who thus might reduce output. A tariff, while also raising the domestic price and reducing imports, would encourage domestic production and the non-equivalance results.

This discussion has assumed that it is possible to choose tariffs or quotas for identifiable commodities. In practice, tariffs or quotas are levied on relatively broad commodity groups. Dixit (1985) illustrates the problem with the choice of policy instruments for a group of commodities like “automobiles,” which is, in fact, composed of different types of vehicles that are less than perfectly substitutable for each other. Ideally, the choice of policy would involve a separate tariff or quota for each type, though the policymaker may be constrained to a choice between a uniform tariff or quota. Since such groups usually involve goods that are reasonable substitutes, the public economics literature suggests, they should be taxed at roughly similar ad valorem rates. If this were not the case, relative prices would differ and would result in large substitutions and quantity changes, with attendant deadweight losses. Thus, uniform ad valorem tariffs for such groups would result in the closest approximation to the best policy. A uniform quota would act as a uniform specific tariff, implying a too-low tariff on commodities with high world prices and a too-high tariff for commodities with low world prices, and would cause a shift in the world supply of goods toward those whose tariffs were too low.

The choice between tariffs and quotas has also been discussed in the context of uncertainty. It is not generally possible to rank tariffs as equivalent or preferable to quotas, and the answers would depend, inter alia, on the nature of the uncertainty and on the slopes of the demand and supply curves (Newbery and Stiglitz (1981)).

Quotas and tariffs also differ in terms of their costs of implementation. The administrative costs of collecting tariffs would vary, depending on the nature of the commodity, the efficiency of the customs officials, and whether there were opportunities to evade payment of duties. These administrative arguments are often proposed in favor of quotas when there is a multiplicity of high tariff rates. The incentives to smuggle would exist with quotas and high tariff rates. The real resource cost of smuggling has been examined by Sheikh (1974).

Krueger (1974) discusses the “rent-seeking” costs involved in the imposition of tariffs and quotas. On the one hand, real resources will be expended to capture the monopoly rents conferred by the quotas. This may be done through competition between potential licensees or bribery of the license-giving agencies. On the other hand, tariffs also generate protection for domestic producers, and there are powerful pressure groups which lobby for the continuation or extension of this protection. Bhagwati and Srinivasan (1980) extended the concept of rent-seeking to the case of “revenue-seeking,” arguing that since tariffs are a fruitful source of revenue, this would generate a lobby for the perpetuation of the benefits obtained in the same way that individuals indulge in rent-seeking activities. Such profit-seeking activities do not generate commodities for consumption, but rather lead to income transfers.

2. Protection

The concepts of effective tariffs and effective protection have come to be widely used. Where intermediate goods are involved, the extent to which an industry is protected is taken to depend not only on the protection accorded to the output but also on that accorded to the inputs, since output tariffs increase the domestic price above world levels and make production more profitable, and input tariffs make production of output more costly, and hence less profitable. A common definition of effective protection is the difference between value added at domestic prices and value added at world prices as a proportion of value added at world prices. This is a positive concept that involves an attempt to measure the resource pulls resulting from trade restrictions. However, it is not, in general, true that resources will flow toward a sector if its effective protection increases. (See Jones and Neary (1984).) Attempts have been made to salvage the “effective protection” idea by imposing strong restrictions on the technology (fixed coefficients or separability between intermediate goods and factors of production), or to suggest formulas which might predict gross output changes. If technical coefficients are not fixed, it is not clear whether the relevant coefficients to be used in the effective-protection calculations should be based on world or domestic prices; they are functions of relative prices of all inputs and also involve nominal tariffs in the calculation of domestic prices. Moreover, gross output levels in conventional welfare analysis “are of no concern. What matters is the net production that is available to consumers, “9 and the nominal tariff rate is the appropriate measure for this. These difficulties render the concept far from useful as a normative measure, although effective protection as a descriptive tool describes ex post the difference between foreign exchange earned and factor payments made in terms of border prices or trade opportunity costs.

Effective protection calculations do not, in fact, guide policy toward sectors that should be protected or indicate which activities should be contracted. The analysis of sectoral policy essentially involves information on the opportunity costs of nontraded goods and factors. “In this regard, the domestic resource cost measure will appropriately reflect the opportunity cost of tradable-goods production. The effective tariff measure, by ignoring the indirect cost of home goods, does not.”10 Apart from its treatment of income distribution and factors, the domestic-resource-cost method is like that used in the Little-Mirrlees shadow-pricing calculations. (See Section IV.) Note that effective protection measures, domestic resource costs, and shadow prices all adjust domestic market prices to reflect trade opportunity costs; however, shadow prices provide the most general guide to policy formulation.

It may be argued that from the normative point of view, protection of any kind is irrelevant. Using the principle of targeting, the best way is to institute subsidies to factors that do not cause by-product distortions and to not use protection at all. Dixit (1985) dismisses the criticism by laymen that

import tariffs raise revenue, while production subsidies cost money and put additional strain on other uses of the government’s budget. We see this argument to be fallacious. The size of the government’s budget has no direct welfare relevance. The optimality conditions show that, on considering the overall effects, it is desirable to provide these production subsidies and adjust other taxes appropriately. 11

This may be seen intuitively if one considers the fact that a tariff is equivalent to a production subsidy and consumption tax, and thus is just one “way of financing (in fact overfinancing) the subsidy.”12 Thus, the arguments for using tariffs to provide protection are largely political or administrative—suggesting that it might be easier to levy a tariff than to provide a production subsidy—though both methods would be open to rent-seeking activities.

3. Export Promotion

A number of export-promotion measures were extant in India in the late 1970s, and not all of them could be described as subsidies, since they were designed to compensate for some of the biases against exports introduced elsewhere. These measures included, inter alia, (i) a duty-drawback scheme; (ii) cash compensatory support and rebates for market development; and (iii) an import-replenishment scheme. Similar instruments have been used in the other countries of the subcontinent. This paper will, however, concentrate on the Indian case.

The duty-drawback scheme refunds duties paid on inputs used to produce exported goods. Import duties are generally refunded in full, although this does not apply to capital goods. Excise duties are also refunded to the extent these can be related to identifiable inputs. In 1975/76, 62 percent of duty-drawback rates were less than 10 percent of the f.o.b. value of the exported goods, and 89 percent of drawbacks were less than 20 percent of the f.o.b. price. The commodities affected covered 97 percent of the cumulative f.o.b. value of exports.13

It is clear that the duty-drawback scheme only partially corrects for the taxation of intermediate goods in the economy, since only some of the directly identifiable taxation of inputs is rebated. The cash-compensatory-support scheme provided assistance in the export of nontraditional products to compensate for unrefunded taxes and levies paid on exported goods and their inputs through the production process. The percentage of cash compensatory support with respect to the f.o.b. value of exports is shown in Table 1 for much of the 1970s, and Table 2 shows that most of the exports receiving cash compensatory support did so in the 10–15 percent range. An interest subsidy is paid out of the market development assistance fund, which also covers the cash-compensatory-support scheme. In 1976/77, this amounted to Rs 100 million out of the market development assistance fund of Rs 2,400 million.

The import-entitlement scheme, which was of considerable importance before the devaluation of the rupee in 1966, allowed exporters to retain a part of the foreign exchange they earned, which entitled them to import twice their input content. These licenses carried a substantial market premium. (See Bhagwati and Desai (1970).) The import-replenishment scheme, which replaced the entitlement scheme, licensed the full import content of the exports of specified goods for import replenishment. The changes made in this scheme through the late 1970s are described in Bagchi (1981). In 1977/78, the value of import licenses issued against exports was Rs 7,410 million, as against Rs 4,670 million on account of the duty-drawback and market-development schemes (including cash compensatory support).

Table 1.India: Cash Compensatory Support Relative to Exports (f.o.b.)(In percent)
Simple

Average

for period
Difference Between

Minimum and

Maximum for
Products1970/711974/751975/761976/77to 1976/771970/71–1976/77
(1)(2)(3)(4)(5)(6)(7)(8)
1.Engineering products14.516.114.815.315.62.20
2.Chemical and allied products; paper products13.714.413.312.714.162.60
3.Plastic goods9.49.29.710.19.762.10
4.Sports goods14.519.416.214.917.645.40
5.Woollen carpets; rugs and druggets10.55.711.015.110.279.40
6.Processed food items8.58.013.812.89.916.40
7.Woollen blended knitwear15.012.113.552.90
8.Fish and fish preparations6.14.95.51.20
9.Instant teas; packet teas and tea bags10.09.99.950.10
10.Jute manufactures9.89.29.50.6
11.Rayon and synthetics14.913.914.41.0
12.Finished leather and leather manufactures5.065.26.16.53.30
13.Natural silk fabrics, garments, and made-ups7.069.610.99.53.30
14.Instant coffee extracts and essence10.0
15.Walnut kernels and walnuts in shell5.018.712.18.637.0
16.Iron and steel scrap (ferrous scrap)5.003.0819.97.216.10
17.Decorticated cottonseed cakes12.0610.0810.018.614.158.60
18.Groundnut cake extractions3.024.537.20
19.Prime iron and steel14.025.94.912.1813.00
20.Iron ore and manganese ore0.054.776.70
21.Machine-twisted curled coir fiber and carpets14.034.075.16.39.449.40
22.Rice bran (extractions)1.007.37.3312.70
Table 2.India: Classification of Exported Items Subject to Cash Compensatory Support at Different Rates, 1976/771
Number of Exported Items
Products<10 percent
10 percent

and

<15 percent
15 percent and

<20 percent
20 percent
Total

Number
Engineering products61337037246
Chemicals and allied products2179255211
Plastics3131
Leather1427
Sports goods11
Fish products1113
Fresh and processed foods22711949
Carpets and handicrafts11114
Woollen fabrics and made-ups33
Natural silk fabrics and garments246
Synthetic fabrics and garments2212346
Coir producrs33
Decorticated cottonseed extractions448
Decorticated cottonseed expeller cake448
Cotton bagging33
Miscellaneous group516
Total2140814858635

In Table 2, the cash compensatory support (CCS) for the export of some of the goods shown in Table 1 may be compated with estimates of the tax element in the price of final goods that arises from the taxation of intermediate goods and raw materials. The effective tax, te defined as the total tax component of the price of final goods, is calculated using actual revenue collections for India for 1979/80. (For details, see Ahmad and Stern (1983, 1987).) This takes into account the taxation of inputs, of inputs into inputs, and so on using a transaction-flow table based on the assumption of full forward shifting of taxes. At the sectoral level of classification of the input-output table, this is not an unreasonable assumption, although for individual commodities whose prices are fixed by the world market, say, changes in input taxes affect payments to factors rather than the final price. The taxation of a good through the taxation of inputs, tdiff, is the difference between te and the nominal tax. Table 3 presents estimates from Ahmad and Stern (1987) for 89 input-output sectors for 1979/80.

Although the element of taxation arising through capital goods is not included in the Ahmad and Stern estimates, tdiff is of the order of 10–15 percent of the producer price of most of the exportable-goods categories, and is as high as 24 percent for “plastics and synthetic rubber goods,” 32 percent for “woollen and silk textiles,” and 44 percent for artificial “silk textiles.” Compare these figures with the cash-compensatory-support estimates for immediately preceding years (Table 1). Although the commodity classifications are not quite identical, a number of commodities appear to have CCS levels considerably lower than tdiff. plastic goods, knitwear, rayon and synthetics; and commodities such as engineering goods, chemicals, and iron and steel products have CCS levels roughly in line with the taxation of intermediate goods. In a number of cases, however, the CCS level exceeds both the effective tax and the tax element arising from the taxation of intermediate goods: these include carpets, leather products (though not footwear), and processed food items. In other cases, the CCS level lies between te and tdiff.

While export incentives have been used to correct for distortions created by policy for protection and for revenue, export subsidies have also been given, as was done under the Pakistan Bonus Voucher Scheme (see Islam (1981)) prior to the 1972 devaluation, to correct for an overvaluation of the currency. In this case, the right policy, which was followed by both the Indian and Pakistan Governments in the recent past, is not to maintain extremely overvalued exchange rates.

Table 3.India: Taxation of Inputs
CommodityietdifftdiffCtdiffSProportion of

Gross

Domestic

Output
1.Rice and products-0.035-0.009-0.0140.0060.0445
2.Wheat and products0.069-0.018-0.0270.0080.0226
3.Jowar and products0.0120.0020.0010.0020.0075
4.Bajra and products0.0030.0040.0020.0050.0025
5.Other cereals0.0090.0130.0020.0110.0066
6.Pulses0.0480.0380.0220.0160.0108
7.Sugarcane0.003-0.010-0.0150.0050.0135
8.Jute-0.001-0.003-0.0040.0010.0010
9.Cotton-0.0050.0050.0000.0050.0077
10.Plantations-0.014-0.016-0.0430.0270.0060
11.Other crops0.005-0.001-0.0030.0020.0617
12.Milk and products0.0090.008-0.0030.0110.0308
13.Other animal husbandry products0.0140.014-0.0040.0180.0157
14.Forestry and logging0.0510.0120.0070.0050.0070
15.Fishing0.0120.0120.0070.0050.0042
16.Coal and lignite0.0650.0180.0090.0080.0060
17.Petroleum and natural gas0.2010.0360.0270.0090.0023
18.Iron ore0.2330.1340.1040.0300.0006
19.Other minerals0.0580.0730.0560.0170.0020
20.Miscellaneous food products0.0670.0470.0210.0260.0178
21.Sugar0.2020.0380.0160.0220.0088
22.Gur and khandsari0.0590.0420.0040.0380.0107
23.Vanaspati0.1910.0750.0390.0360.0064
24.Other edible oils0.0740.015-0.0080.0240.0062
25.Tea and coffee0.2210.0500.0200.0300.0044
26.Other beverages3.5900.0740.0490.0250.0012
27.Tobacco manufactures0.9390.1150.0970.0150.0047
28.Cotton textiles excluding handloom and khadi0.1080.0510.0340.0170.0248
29.Cotton textiles, handloom and khadi0.0700.0820.0560.0260.0122
30.Woollen and silk textiles0.4110.3250.2970.0280.0026
31.Artificial silk fabrics0.5980.4420.4320.0100.0035
32.Jute textiles0.1420.0690.0530.0160.0045
33.Readymade garments0.0930.0810.0670.0130.0084
34.Miscellaneous textile products0.1320.1140.0910.0230.0065
35.Carpet weaving0.0520.0480.0380.0110.0015
36.Wood products0.0930.0780.0570.0210.0066
37.Paper, products and newsprint0.2750.1060.0750.0310.0061
38.Printing and publishing0.0900.0900.0680.0220.0046
39.Leather and products0.0530.0540.0270.0270.0033
40.Leather footwear0.1830.1220.1010.0210.0023
41.Rubber products0.4080.1610.1240.0370.0065
42.Plastics and synthetic rubber0.4800.2480.2160.0320.0036
43.Petroleum products0.5480.0130.0090.0040.0153
44.Miscellaneous coal and petroleum products0.5180.0870.0650.0220.0029
45.Inorganic heavy chemicals0.2620.0790.0600.0190.0039
46.Organic heavy chemicals0.3620.1270.1030.0240.0010
47.Chemical fertilizers-0.2350.0420.0170.0250.0102
48.Insecticides, fungicides, etc.0.2410.1730.1190.0530.0010
49.Drugs and pharmaceuticals0.3000.1660.1160.0500.0096
50.Soaps and glycerines0.2670.1190.0720.0470.0024
51.Cosmetics0.2600.0860.0500.0360.0020
52.Synthetic rubber and fibers1.0290.1310.0990.0310.0029
53.Other chemicals0.3880.1740.1430.0310.0047
54.Refractories0.1020.0490.0350.0130.0037
55.Cement0.4570.0790.0580.0220.0024
56.Other nonmetallic products0.1480.0960.0740.0220.0101
57.Iron and steel, ferroalloys0.1340.1130.0930.0200.0157
58.Castings and forgings0.1100.1100.0930.0170.0013
59.Iron and steel structures0.5570.1410.1230.0180.0035
60.Nonferrous metals, alloys0.1710.1010.0840.0170.0106
61.Metal products0.1700.1140.0980.0170.0147
62.Tractors and agricultural implements0.2200.0890.0560.0340.0024
63.Machine tools0.2280.1250.0990.0260.0015
64.Office, domestic, and commercial equipment0.2280.0730.0550.0180.0007
65.Other non-electrical machinery0.1630.1340.1200.0140.0151
66.Electric motors0.2670.1250.1000.0250.0024
67.Electrical cables and wires0.3310.1510.1290.0220.0033
68.Batteries0.3370.0870.0720.0150.0010
69.Household electrical goods0.3570.1120.0910.0210.0011
70.Communications and electronic equipment0.2390.1370.1150.0220.0022
71.Other electrical machinery0.2190.1290.1060.0230.0065
72.Ships and boats0.0700.0700.0540.0160.0008
73.Rail equipment0.1070.1070.0790.0280.0046
74.Motor vehicles0.1760.1180.0890.0290.0088
75.Motorcycles and bicycles0.1570.1140.0930.0210.0035
76.Other transport equipment0.1300.0860.0620.0240.0005
77.Watches and clocks0.3580.1510.1150.0360.0004
78.Miscellaneous manufacturing industries0.3600.1110.0940.0170.0121
79.Construction0.0650.0650.0430.0220.0695
80.Electricity, gas, and water supply0.1130.0730.0550.0180.0202
81.Railways0.0520.0520.0370.0150.0127
82.Other transport0.1500.1150.0810.0340.0438
83.Communications0.0110.0110.0070.0040.0048
84.Trade, storage, warehouses0.0100.0180.0100.0040.0907
85.Banking and insurance0.0040.0060.0030.0020.0184
86.Real estate, owner dwellings0.0070.0080.0050.0030.0216
87.Education0.0920.0930.0660.0270.0451
88.Medical health0.2100.2100.1170.0930.0186
89.Other services0.0460.0400.0260.0140.0593
Source: Ahmad and Stern (1987).Note: tx = effective taxes from all sources as a proportion of the producer pricetdiff= tet, where tx denotes the overall effective tax vector and t denotes the nominal vector of commodity taxestdiffC= tdiffCtC, where txC denotes the effective tax from union taxes: excises, import duties, and subsidies, with t the corresponding nominal tax vectortdiff$= tet$, where tx$ denotes the effective state tax vector, comprising sales tax, state excises, and other taxes, and tS denotes the equivalent nominal tax vector.The proportion of sectoral gross value of output in total domestic output is shown in the column on the far right.

Whether export subsidies through CCS and the duty drawbacks are sufficient to provide a substantial impetus to exports cannot be determined, except on a case-by-case basis which examines not only the determinants of domestic supply but also the relationship between the ex-factory and world prices, and the foreign sources of demand. If there are costs of entry into established markets, including explicit quotas and import barriers imposed by third countries, these would be subsumed under the latter. The Dagli Committee (India, Ministry of Finance (1979)) argued that although there had been a significant increase in India’s exports since 1972/ 73 (shown in Table 4),

assisted exports showed a really significant increase only in the last two years, namely, 1975—76 and 1976—77, when the simple average of yearly growth rates of such exports was as much as 54 per cent, against only 16 per cent for non-assisted exports.14

Bagchi (1981) has questioned this view, and that of the Alexander Committee (India, Ministry of Commerce (1978)), on the grounds that the subsidies were, in fact, larger than was assumed; that a breakdown of the commodities covered under the CCS suggested that only 6 out of 22 groups had export growth rates higher than the average growth of exports of commodities not covered; and that a comparison of growth rates per se cannot lead to any firm conclusions. The Dagli Committee (India, Ministry of Finance (1979)) also correctly cautions against the duty drawbacks and export subsidies in the case of commodities in which India enjoys monopoly power, since these would only serve to turn the terms of trade against the country. In general, however, the small-country assumption is valid for most goods, and rebating the taxation of inputs merely removes the domestic bias against exports and does not necessarily imply that all commodities could, or should, be exported.

Table 4.Cash Compensatory Support (CCS) and Duty Drawback (DD) on Exports from India over 1970–71 to 1976–77
Yearf.o.b. Value of Total Exports from Indiaf.o.b. Value of Exports on Which CCS Existsf.o.b. Value of Exports of Items Not Having CCSValue of CCS Paid to ExportersValue of DD Allowed to ExportersTotal Value of CCS and DD Received by Exporters(3)

(2)
(5)

(3)
(7)

(3)
Yearly Growth Rate of Total Exports

(2)
Yearly Growth Rate of Assisted Exports

(3)
Yearly Growth Rate of Non- Assisted Exports

(4)
(1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)
1970–7115,3503,38011,98041033074022.012.121.9
1971–7216,0803,55012,54054036090022.116.125.74.85.04.7
1972–7319,7104,38015,3306204701,09022.214.125.022.623.422.3
1973–7424,8305,68019.1506204301,05022.910.918.526.029.824.9
1974–7533,2906,50026,7907406001,34019.511.720.634.014.439.9
1975–7640,43010,62029,8101,4808202,30026.313.921.721.463.411.3
1976–7751,34015,27036,0702,4001,0003,40029.715.722.327.043.821.0
Source: India, Ministry of Finance (1979).

Although most taxes are liable to be evaded to some extent, the rebating of a complex structure of excises and import duties is not simple. These rebates are liable to much abuse and rent-seeking activity, and there is much anecdotal evidence that this has been the case in both India and Pakistan. In Western Europe, in particular, the principle of “border tax adjustment” permits the refund or all indirect taxes involved in the production of export goods through the zero-rating of exports under the value-added tax (VAT). A partial VAT on manufacturing (MANVAT) was suggested for India by the Indirect Taxation Reforms Committee (the Jha Committee (India, Ministry of Finance (1978))), and a modified version has recently been introduced. A major constraint on the adoption of a full-scale VAT in India is the constitutional division of taxation powers between the federal government and the states, with the former responsible for sales taxes and the latter for customs and excises. Authorities in other countries are not so constrained, and more than 20 developing countries have now adopted variants of the VAT (Tanzi (1987)). A recent report on tax reform in Pakistan has also recommended a VAT to overcome the problems of the taxation of inputs, the unintended protection of some sectors, and the bias against exports and the rent-seeking activities associated with ad hoc tax rebates. (See Ahmad and Stern (1986 b).)

Whether there should be subsidies additional to the rebate of taxes depends on arguments which parallel the discussions relating to infant industries and which are subject to similar objections. These could also be analyzed in a game-theoretic context of retaliatory policies by trading partners and several “beggar-thy-neighbor” results can obtain. Some subsidies run afoul of General Agreement on Tariffs and Trade (GATT) rules, and accusations of dumping reinforce protectionism around the world, even among the richer OECD nations. In any case, if the social costs of export promotion exceed the social benefits, perhaps the country would be better advised to follow alternative policies.

IV. Shadow Prices and Policy

The use of shadow prices as a guide to policy and reform has been discussed briefly in Section II. Empirical estimates for India based on the Planning Commission 1979–80 inter-industry flows matrix were presented in Ahmad, Coady, and Stern (1986). Shadow prices of tradables are determined with reference to border prices. The shadow price for nontradables, in terms of the Little-Mirrlees method, is the marginal cost valued at shadow prices of an extra unit of production (discussed in Dreze and Stern (1985)), and this involves knowing the input requirements and the shadow prices of these inputs. These inputs include traded goods, nontraded goods, and factors of production; and input-output methods are used in the estimation. Table 6 in the Appendix sets out schematically the determination of endogenous shadow prices for nontradables and exportables. (The latter are endogenous in this case, since the input-output table is formulated in terms of producer prices, and nontraded margins are involved in getting the goods to the border.) The shadow prices for importables and factors are exogenously determined. (For details concerning method and limitations, see Ahmad, Coady, and Stern (1986).)

For India, different sets of shadow prices were calculated, corresponding to various assumptions concerning the classification of goods into tradables and nontradables and the valuation of factors. This corresponds to different policy options concerning trade, labor markets, and the treatment of capital goods and factors, and encompasses a range of plausible scenarios for the Indian economy. (For applications to the Pakistan economy, see Ahmad, Coady, and Stern (1984).) Table 7 in the Appendix sets out the classification of sectors into importable, exportable, and nontraded. In Case 1, 32 sectors are treated as importable at the margin, 39 as exportable, and 18 as nontradable. Case 2 treats some of the agricultural and service sectors taken as tradable in Case 1 as nontradable, and there are 36 exportable, 27 importable, and 26 nontradable sectors. In Case 3, additional manufacturing sectors are treated as nontradable, and there are 47 nontradable, 27 exportable, and 15 importable sectors. The different classifications are influenced partly by the possibility that there may be a policy change, since a relaxation of an import quota might change a good’s status from nontradable to tradable, and partly by the heterogeneity of some of the sectors, which could include, for example, commodities which are importable as well as those produced domestically on the margin. Thus, a movement from Case 3 toward Case 1 could be seen as an adjustment from a restrictive to a more traded environment.

A further degree of sensitivity is introduced in the treatment of the shadow prices of factors, which represent the opportunity cost in terms of social welfare of the employment of, and the payment to, each factor. The market price of the factor can be converted to a shadow price for the type of good the factor might produce, using a standard conversion factor (SCF). Given that the numeraire is foreign exchange, the SCF may be seen as the reciprocal of the “shadow exchange rate.” Alternative values have been chosen for the conversion factors for assets and labor. These are 0.75, 0.50, and 0.25 for assets; and 0.90, 0.75, and 0.50 for labor. This provides a “plausible” range for a sensitivity to alternative models of the Indian economy relating to capital and labor markets. The value of 0.75 could represent, for instance, an average value for the extent to which domestic prices exceed world prices for tradable goods and could crudely represent the SCF. In this case, using 0.75 for wages and assets would represent the assumption that wages and assumed capital costs (an 8 percent real rate of interest and given sectoral incremental capital-output ratios (ICORs)) reflect opportunity costs at market prices. With unemployed factors, opportunity costs may be lower than market prices, as is true for labor. The different combinations of conversion factors could also reflect changes in the exchange rate, and thus a wide range of policy options could be covered.

The social profitability of the 89 sectors, defined as the shadow profit divided by the shadow value of output, for a conversion factor of 0.75 for assets and the three alternatives for labor (0.90, 0.75, and 0.50) for Case 1 (the more traded option) is shown in Table 5. Nontraded sectors break even by definition. (The other sets of calculations are described in Ahmad, Coady, and Stern (1986).) Despite the sensitivity analysis involved in the widely varying valuations for factors and the alternative classification of sectors from the most restrictive set of trade policies to gradual liberalization, the results from Ahmad, Coady, and Stern (1986) suggest that a number of sectors that appeared to make a commercial loss were socially profitable across some assumptions. These include cotton textiles, jute textiles, leather footwear, metal products, and household electrical goods. On the other hand, some sectors which made a commercial profit were socially unprofitable, including plastics and synthetic rubber, petroleum products, inorganic heavy chemicals, and drugs and pharmaceuticals.

Table 5.Social Returns and Their Sensitivity to the Accounting Ratio (AR) for Labor with the AR for Assets at 0.75 (Case 1)
Labor
0.90.750.5
1.Rice and products0.15940.22330.3281
2.Wheat and products0.15560.21250.3058
3.Jowar and products0.00000.00000.0000
4.Bajra and products0.00000.00000.0000
5.Other cereals0.00000.00000.0000
6.Pulses0.00000.00000.0000
7.Sugarcane0.00000.00000.0000
8.Jute0.16820.24380.3675
9.Cotton0.20300.26890.3774
10.Plantations0.17100.20300.2564
11.Other crops0.22110.28990.4028
12.Milk and products0.08400.13200.2119
13.Other animal husbandry products0.00000.00000.0000
14.Forestry and logging0.31590.39130.5127
15.Fishing0.19670.24110.3137
16.Coal and lignite0.18650.24430.3406
17.Petroleum and natural gas0.22560.28350.3800
18.Iron ore0.45510.48540.5339
19.Other minerals0.21700.27130.3619
20.Miscellaneous food products0.10940.14140.1929
21.Sugar0.16350.23060.3397
22.Gur and khandsari0.00000.00000.0000
23.Vanaspati0.00000.00000.0000
24.Other edible oils0.08010.08940.1051
25.Tea and coffee0.25310.27430.3087
26.Other beverages0.00000.00000.0000
27.Tobacco manufactures0.14670.20050.2852
28.Cotton textiles excluding handloom and khadi0.08590.12670.1930
29.Cotton textiles, handloom and khadi0.13530.17500.2399
30.Woollen and silk textiles0.33270.35550.3930
31.Artificial silk fabrics0.53730.56680.6148
32.Jute textiles0.06760.11280.1861
33.Ready made garments0.18840.24320.3333
34.Miscellaneous textile products0.23580.26510.3128
35.Carpet weaving0.15950.22310.3250
36.Wood products0.26870.31920.4007
37.Paper, products and newsprint0.06400.10530.1741
38.Printing and publishing0.17020.22870.3264
39.Leather and products0.22390.25910.3161
40.Leather footwear0.11530.17120.2609
41.Rubber products0.24410.27880.3346
42.Plastic and synthetic rubber-0.1641-0.1228-0.0540
43.Petroleum products-0.1187-0.1160-0.1116
44.Miscellaneous coal and petroleum products0.18820.21800.2659
45.Inorganic heavy chemicals-0.5450-0.4784-0.3674
46.Organic heavy chemicals-0.3502-0.2965-0.2070
47.Chemical fertilizers0.12420.15680.2112
48.Insecticides, fungicides, etc.0.09570.12180.1654
49.Drugs and pharmaceuticals-0.1356-0.0972-0.0330
50.Soaps and glycerines0.20100.23020.2778
51.Cosmetics0.15020.17060.2039
52.Synthetic rubber and fibers-0.9619-0.8775-0.7368
53.Other chemicals-0.01110.02460.0842
54.Refractories0.15350.19690.2691
55.Cement0.14180.16890.2140
56.Other nonmetallic products0.19470.24600.3281
57.Iron and steel, ferroalloys-0.2098-0.1473-0.0432
58.Castings and forgings0.36070.38780.4316
59.Iron and steel structures0.32080.34780.3929
60.Nonferrous metals, alloys-0.1864-0.1386-0.0590
61.Metal products0.25790.30630.3845
62.Tractors and agricultural implements-0.4144-0.3327-0.1965
63.Machine tools-0.1283-0.0811-0.0024
64.Office, domestic, and commercial equipment-0.2781-0.1820-0.0220
65.Other non-electrical machinery-0.3189-0.2654-0.1761
66.Electric motors-0.3015-0.2569-0.1827
67.Electrical cables and wires0.34390.37890.4348
68.Batteries0.27270.31550.3844
69.Household electrical goods0.21490.26570.3479
70.Communications and electronic equipment-0.4373-0.3614-0.2350
71.Other electrical machinery-0.6996-0.5886-0.4036
72.Ships and boats-0.3514-0.2570-0.0997
73.Rail equipment0.15180.19510.2651
74.Motor vehicles0.19820.23760.3009
75.Motorcycles and bicycles0.21320.25950.3338
76.Other transport equipment0.10640.14650.2132
77.Watches and clocks-0.2563-0.1825-0.0596
78.Miscellaneous manufacturing industries0.22450.29120.3974
79.Construction0.00000.00000.0000
80.Electricity, gas, and water supply0.00000.00000.0000
81.Railways0.00000.00000.0000
82.Other transport0.00000.00000.0000
83.Communications0.00000.00000.0000
84.Trade, storage, warehouses0.00000.00000.0000
85.Banking and insurance0.24740.35160.5251
86.Real estate, owner dwellings0.00000.00000.0000
87.Education0.00000.00000.0000
88.Medical health0.00000.00000.0000
89.Other services0.16710.24900.3854
Source: Ahmad, Coady, and Stern (1986).Notes: (i) The social returns are defined as the shadow profit divided by the shadow value of output.(ii) Nontraded sectors break even.(iii) The classification of sectors is given in Table 2.

In general, agricultural and natural-resource sectors were socially profitable, as was the light-engineering sector, reflecting India’s relative resource endowments. And these might be industries that would qualify for further expansion in either domestic or export markets. However, some of the heavier, mainly capital-intensive sectors displayed negative commercial and social returns, including tractors and agricultural implements, machine tools, electric motors, other non-electrical machinery, and communications equipment. This might indicate either that during the period of infancy, some of these sectors had not succeeded in reducing unit costs of production, or that quotas and licensing resulted in poor capacity utilization, reflecting the fact that “the framework of economic policies governing industrialization does not induce or permit systematic attention to costs.”15 If a case could be made that the heavy manufacturing activities were operating well below their potential levels of efficiency, then policies should be directed toward ameliorating constraints on the efficient use of existing resources. Otherwise, additional investment should be directed toward more socially profitable sectors. And the shadow prices could be used to guide import and licensing policies toward these sectors, rather than operating on a first-come, first-served or an indigenous availability basis.

Thus, the use of shadow prices is helpful in identifying the areas in which adjustments in the economy might proceed with respect to exchange rates, to policy changes reflecting changes from quotas to tariffs, or to other policy measures affecting capital or labor markets. These shadow-price calculations indicate sectors that might be encouraged, but a more detailed analysis at the commodity level would be needed to make the recommendations operational. And the previous sections of this paper have indicated which instruments might be used to carry out the adjustments.

V. Concluding Remarks

This paper has examined the approaches to trade regimes in view of the discussion relating to structural adjustment in South Asia. The literature review of Section II suggests that while there are likely to be gains from trade, neither laissez faire nor autarky emerge as desirable policy options. Section IV illustrates that under a variety of assumptions, there are several sectors in India that should be encouraged. The choice of instruments is seen, in Section III, to be particularly important, since some methods of protection are open to abuse and may have unintended consequences for other sectors of the economy. In particular, the Indian system of input taxation, which was designed to generate revenue and provide protection, has had the effect of “excessively” taxing exportables. This paper also stresses the interrelationship between different aspects of public policy, and it is only when these linkages are recognized that serious policy errors can be avoided.

Appendix

Table 6.Shadow Prices for Nontraded Goods and Exportables

Table 7.India: Classification of Sectors1
SectorCase 1Case 2Case 3
1. Rice and productsXXX
2. Wheat and productsXXN
3. Jowar and productsNNN
4. Bajra and productsNNN
5. Other cerealsNNN
6. PulsesNNN
7. SugarcaneNNN
8. JuteXNN
9. CottonXXX
10. PlantationsMXN
11. Other cropsXNN
12. Milk and productsMXN
13. Other animal husbandry productsNNN
14. Forestry and loggingXNN
15. FishingXNX
16. Coal and ligniteMMM
17. Petroleum and natural gasMMM
18. Iron oreXXX
19. Other mineralsMXM
20. Miscellaneous food productsXXX
21. SugarXXX
22. Gur and khandsariNNN
23. VanaspatiNNN
24. Other edible oilsMMX
25. Tea and coffeeXXX
26. Other beveragesNNX
27. Tobacco manufacturesXXX
28. Cotton textiles excluding handloom and khadiXXX
29. Cotton textiles, handloom and khadiXXX
30. Woollen and silk textilesXXX
31. Artificial silk fabricsXMX
32. Jute textilesXXX
33. Readymade garmentsXXX
34. Miscellaneous textile productsXXX
35. Carpet weavingXXX
36. Wood productsXXX
37. Paper, products and newsprintMMM
38. Printing and publishingMNN
39. Leather and productsXXX
40. Leather footwearXXX
41. Rubber productsXXX
42. Plastic and synthetic rubberMMM
43. Petroleum productsMMM
44. Miscellaneous coal and petroleum productsXXX
45. Inorganic heavy chemicalsMMM
46. Organic heavy chemicalsMMM
47. Chemical fertilizersMMN
48. Insecticides, fungicides, etc.MMM
49. Drugs and pharmaceuticalsMXN
50. Soaps and glycerinesXXX
51. CosmeticsXMN
52. Synthetic rubber and fibersMMM
53. Other chemicalsMMM
54. RefractoriesMMN
55. CementMMN
56. Other nonmetallic productsXXX
57. Iron and steel, ferroalloysMMM
58. Castings and forgingsXXX
59. Iron and steel structuresMXN
60. Nonferrous metals, alloysMXN
61. Metal productsXMN
62. Tractors and agricultural implementsMMN
63. Machine toolsMMN
64. Office, domestic, and commercial equipmentMXN
65. Other non-electrical machineryMXN
66. Electric motorsMMN
67. Electrical cables and wiresXXN
68. BatteriesXXX
69. Household electrical goodsXXX
70. Communications and electronic equipmentMMN
71. Other electrical machineryMXM
72. Ships and boatsMMM
73. Rail equipmentXXX
74. Motor vehiclesXMN
75. Motorcycles and bicyclesXXX
76. Other transport equipmentMMN
77. Watches and clocksMMN
78. Miscellaneous manufacturing industriesXMN
79. ConstructionNNN
80. Electricity, gas, and water supplyNNN
81. RailwaysNNN
82. Other transportNNN
83. CommunicationsNNN
84. Trade, storage, warehousesNNN
85. Banking and insuranceXNN
86. Real estate, owner dwellingsNNN
87. EducationNNN
88. Medical healthNNN
89. Other servicesXNN
Source: Ahmad, Coady, and Stern (1986).
References

    Ahmad, Ehtisham, DavidCoady, and NicholasStern, “Tax Reform, Shadow Prices, and Effective Taxes: Illustrations for Pakistan for 1975–76” (unpublished, Development Economics Research Center, University of Warwick, England, July1984).

    Ahmad, Ehtisham, DavidCoady, and NicholasStern, — “Shadow Prices and Commercial Policy for India 1979/80” (unpublished, Development Economics Research Center, University of Warwick, England, June1986).

    Ahmad, Ehtisham, and NicholasStern, “Effective Taxes and Tax Reform in India” (unpublished, Development Economics Research Center, University of Warwick, England,January1983).

    Ahmad, Ehtisham, and NicholasStern, “Theory of Reform and Indian Indirect Taxes,”Journal of Public Economics (Amsterdam), Vol. 25 (December1984), pp. 259-98.

    Ahmad, Ehtisham, and NicholasStern, (1986 a), “Tax Reform of Pakistan: Overview and Effective Taxes for 1975-76,Pakistan Development Review (Islamabad), Vol. 25Spring1986), pp. 43-72.

    Ahmad, Ehtisham, and NicholasStern, (1986 b), “Analysis of Tax Reform for Developing Countries: Lessons from Research on Pakistan and India” (unpublished, Development Research Program, London School of Economics, December1986).

    Ahmad, Ehtisham, and NicholasStern, “Alternative Sources of Government Revenue, Examples from India, 1979-80,” in The Theory of Taxation for Developing Countries,ed. byDavid M.G.Newbery and NicholasH. Stern (New York: Oxford University Press, 1987), pp. 281-332.

    Bagchi, Amaresh, “Export Incentives in India: A Review,”inChange and Choice in Indian Industry,ed. byAmiyaK. Bagchi and NirmalaBanerjee (Calcutta: K.P. Bagchi, 1981), pp. 297-327.

    Balassa, Bela, “Reforming the System of Incentives in Developing Countries,”World Development (Oxford, England),Vol. 3 (June1975), pp. 365-82.

    Bell, Martin, BruceRoss-Larson, and Larry E.Westphal, “Assessing the Performance of Infant Industries,”Journal of Development Economics (Amsterdam), Vol. 16 (October1984), pp. 101-28.

    Bhagwati, Jagdish N., “Immiserizing Growth: A Geometrical Note,”Review of Economic Studies (Edinburgh), Vol. 25 (June1958), pp. 201-205.

    Bhagwati, Jagdish N., “On the Equivalence of Tariffs and Quotas,”in Trade, Growth, and the Balance of Payments: Essays in Honor of Gottfried Haberler,ed. byRobertE.Baldwin and others (Chicago: Rand McNally, 1965), pp. 5367.

    Bhagwati, Jagdish N., andPadmaDesai, India: Planning for Industrialization: Industrialization and Trade Policies Since 1951 (London: Oxford University Press, 1970).

    Bhagwati, Jagdish N., and T. N.Srinivasan, India, Foreign Trade Regimes and Economic Development,Vol. 6 (New York: National Bureau of Economic Research, 1975).

    Bhagwati, Jagdish N., and T. N.Srinivasan, “Revenue Seeking: A Generalization of the Theory of Tariffs,“journal of Political Economy (Chicago), Vol. 88 (December1980), pp. 1069-87.

    Bhagwati, Jagdish N., and T. N.Srinivasan, Lectures on International Trade (Cambridge, Massachusetts: MIT Press,1983).

    Bhagwati, Jagdish N., and T. N.Srinivasan, and HenryWan, Jr., “Value Subtracted, Negative Shadow Prices of Factors in Project Evaluation, and Immiserizing Growth: Three Paradoxes in the Presence of Trade Distortions,”Economic journal (London), Vol. 88 (March1978), pp. 121-25.

    Birnberg, Thomas B., and StephenA. Resnick, Colonial Development: An Econometric Study (New Haven, Connecticut: Yale University Press, 1975).

    Bliss, C., “Trade and Development: Theoretical Issues and Policy Implications,”in Handbook of Development Economics,ed. byHollisChenery and T.N.Srinivasan (Amsterdam: North-Holland, forthcoming).

    Brecher, Richard A., and JagdishN. Bhagwati, “Foreign Ownership and the Theory of Trade and Welfare,”journal of Political Economy (Chicago), Vol. 89 (June1981), pp. 497-511.

    Chakravarty, Sukhamoy, “Aspects of India’s Development Strategy for the 1980s,”Economic and Political Weekly (Bombay), Vol. 19 (May19, 1984), pp. 845-52.

    Corden, W. Max, “The Normative Theory of International Trade,” inHandbook of International Economics,Vol. 1, ed. byRonaldW. Jones and PeterB. Kenen (Amsterdam: North-Holland, 1984), pp. 63-130.

    Diamond, Peter A., and JamesA. Mirrlees, “Optimal Taxation and Public-Production I: Production Efficiency,”American Economic Review (Nashville), Vol. 61 (March1971), pp. 8-27.

    Diamond, Peter A., and JamesA. Mirrlees, “Optimal Taxation and Public Production II: Tax Rules,”American Economic Review (Nashville), Vol. 61 (June1971), pp. 261-78.

    Dixit, Avinash K., “International Trade Policy for Oligopolistic Industries,”Economic Journal (London), Vol. 94 (Supplement, 1984), pp. 1-16.

    Dixit, Avinash K., “Tax Policy in Open Economies,” inHandbook of Public Economics,Vol, I,ed. byAlanJ. Auetbach and MattinS. Feldstein (Amstetdam: North-Holland, 1985), pp. 313-74.

    Dixit, Avinash K., andVictorD. Norman, Theory of International Trade: A Dual, General Equilibrium Approach (Welwyn, England: J. Nisbet, 1980).

    Dixit, Avinash K., and NicholasStem, “Oligopoly and Welfare: A Unified Presentation with Applications to Trade and Development,”European Economic Review (Amsterdam), Vol. 19 (September1982), pp. 12343

    Drèze, Jean, and NicholasStern, “The Theory of Cost-Benefit Analysis” (unpublished, Development Economics Research Center, University of Warwick, England,March1985). This article is to be published in Handbook of Public Economics,Vol, II,ed. byAlanJ. Auerbach and MartinS. Feldstein (Amsterdam: North-Holland, forthcoming).

    Fields, Gary S., “Export-Led Growth and Labor Markets” (unpublished, Development Economics Research Center, University of Warwick, EnglandJune1983).

    Fields, Gary S., “Employment, Income Distribution and Economic Growth in Seven Small Open Economies,”Economic Journal (London), Vol. 94 (March1984). pp. 74-83.

    Findlay, Ronald, “Growth and Development in Trade Models,” inHandbook of international Economics,Vol. I, ed. byRonaldW. Jones and PeterB. Kenen (Amsterdam: North-Holland, 1984), pp. 185-236.

    Helpman, Elhanan, and AssafRazin, A Theory of International Trade Under Uncertainty (New York: Academic Press, 1978).

    India, Ministry of Commerce, Report of the Committee on Import-Export Policies and Procedures (New Delhi, 1978).

    India, Ministry of Commerce, Ministry of Commerce, Report of the Committee on Trade Policies (New Delhi, 1984).

    India, Ministry of Commerce, Ministry of Finance, Report of the Indirect Taxation Enquiry Committee (Delhi: Controller of Publications, 1978),

    India, Ministry of Commerce, Ministty of Finance, Report of the Committee on Controls and Subsidies (New Delhi, 1979).

    India, Ministry of Commerce, Ministry of Finance, Economic Survey 1985-86, (New Delhi, 1986). Islam, Nurul, Foreign Trade and Economic Controls in Development: The Case of United Pakistan (New Haven, Connecticut: Yale University Press, 1981).

    Johnson, Harry G., “The Possibility of Income Losses from Increased Efficiency or Factor Accumulation in the Presence of Tariffs,”Economic journal (London), Vol. 77 (March1967), pp. 151-54.

    Jones, Ronald W., and J. PeterNeary, “The Positive Theory of International Trade,”in Handbook of international Economics,Vol. I, ed. byRonaldW. Jones and PeterB. Kenen (Amsterdam: North-Holland, 1984), pp. 1-62.

    Krueger, Anne O., “Evaluating Restrictionist Trade Regimes: Theory and Measurement,”Journal of Political Economy (Chicago), Vol. 80 (January/February1972), pp. 48-62.

    Krueger, Anne O., “The Political Economy of the Rent-Seeking Society,”American Economic Review (Nashville), Vol. 64 (June1974), pp. 291-303.

    Krueger, Anne O., “Trade Policies in Developing Countries,”in Handbook of International Economics,Vol. I,ed. byRonaldW. Jones and PeterB. Kenen (Amsterdam: North-Holland, 1984), pp. 519-69.

    Krueger, Anne O., “Trade Policies in Developing Counrries,”in Handbook of International Economics,Vol. I, ed. byRonaldW. Jones and PeterB. Kenen (Amsterdam: North-Holland, 1984), pp. 519-69.

    Krueger, Anne O., and BaranTuncer, “An Empirical Test of the Infant Industry Argument,”American Economic Review (Nashville), Vol. 72 (December1982), pp. 1142-52.

    Lal, Deepak, Prices for Planning: Towards the Reform of Indian Planning (London: Heinemann, 1980).

    Lewis, Stephen R., Taxation for Development: Principles and Applications (New York: Oxford University Press, 1984).

    Lewis, W. Arthur, “The Slowing Down of the Engine of Growth,”American Economic Review (Nashville), Vol. 70 (September1980), pp. 55564.

    Little, I.M.D., and J.A.Mirrlees, Project Appraisal and Planning for Developing Countries (London: Heinemann, 1974).

    Newbery, David M.G., and JosephE. Stiglitz, The Theory of Commodity Price Stabilization: A Study in the Economics of Risk (Oxford, England: Clarendon Press, 1981).

    Newbery, David M.G., and JosephE. Stiglitz, “Pareto Inferior Trade,”Review’ of Economic Studies (Edinburgh), Vol. 51 (January1984), pp. 1-12.

    Ohyama, Michihiro, “Trade and Welfare in General Equilibrium,”Keio Economic Studies (Tokyo), Vol. 9 (1972), pp. 37-73.

    Reidel, James, “Trade as the Engine of Growth in Developing Countries, Revisited,”Economic Journal (London), Vol. 94 (March1984), pp. 5673.

    Sheikh, Munir A., “Smuggling, Production and Welfare,“Journal of International Economics (Amsterdam), Vol. 4 (November1974), pp. 355-64.

    Smith, M.A.M., “Intertemporal Gains from Trade,”Journal of International Economics (Amsterdam), Vol. 9 (May1979), pp. 239-48.

    Tanzi, Vito, “Quantitative Characteristics of the Tax Systems of Developing Countries,”in The Theory of Taxation for Developing Countries,ed. byDavidM.G. Newbery and NicholasH. Stern (New York: Oxford University Press, 1987), pp. 205-41.

Dixit (1985), p. 314.

W. Arthur Lewis (1980), p. 555.

Fields (1983), p. 17.

India, Ministry of Finance (1979), p. 87.

Dixit (1985), p. 361.

Ibid.

India, Ministry of Finance (1979), p. 91.

    Other Resources Citing This Publication