L Policy, Research, and External Affairs WORKING PAPERS Trade Policy Country Economics Department The World Bank January 1991 WPS 569 Domestic Purchase Requirements for Import License Allocations in Mali Wendy E. Takacs To obtain import licenses for sugar and tea, prospective importers in Mali were required to purchase a certain amount of domestic output. The efficiency of this kind of arrangement relative to that of a direct trade restriction such as a tariff depends on the policy's objective and whether the protected industry is competitive or a monopoly. The Policy, Research, and External Affairs Complex disributes PRE Working Papers to disseminate the findings of work in progress and to encourage the exchange of ideas among Bank staff and all others interested in development issues. These papers carry the names of the authors, reflect only their views, and should be used and cited accordingly the findings, interpretations, and conclusions are the authors own. They should not be attributed to the World Bank, its Board of Direc"-m, its management, or any of its member countnes. Policy, Research, and External Affai Trade Policy WPS 569 This paper-- a product of the Trade Policy Division, Country Economics Department- was prepared as background material for the joint UNDP/World BankTrade Expansion Progran, which provides technical and policy advice to countries wishing to reform their trade regimes. Copies are available free from the World Bank, 1818 H Street NW, Washington DC 20433. Please contact Sheila Fallon, room NIO-035, extension 37947 (25 pages, including figures). The govemment of Mali, as part of its trade If the protected industry is competitive, liberalization program, substituted an import linking arrangements are less costly than tariffs licensing system for the state trading agency that if the objective is to increase domestic produc- had held import monopolies on a number of tion or maintain a given degree of self-suffi- products. To get licenses to import sugar and ciency (defined as maintaining a particular ratio tea, prospective importers were required to of imports to domestic production). The reversc purchase given amounts of domestic output. The may be true if the protected market is a mo- volume of imports was thus "linked" to domestic nopoly. output of the imported products. Price controls on the products subject to Takacs investigates the economic impact of linking arrangements dilute the effectiveness of these linking arrangements under two domestic the arrangements and cause disequilibrium in the market structures: perfect competition and market between distributors and consumeis. monopoly. The arrangements have an effect Removing price controls before the linking equivalent to that of a tariff when the tariff arrangements in a liberalization program would revenue is transferred to the domestic industry as drive up prices for both consumers and produc- a subsidy. The cost of these linking arrange- ers, provide false signals, and possibly increase ments to the country imposing them may be the costs of adjusting to liberalization. greater or smaller than the cost of tariffs, de- pending on the policy's objective and the struc- ture of the protected industry. The PRE Working Ilaper SCriCs disseminates the findings of work tnder way in the Bank's Policy, Research, and External Affairs Complex.An objective of the series is to getthese findings out quickly, even if presentations are less than fully polished. The findings, interpretations, and conclusions in these papers do not necessarily represent official Bank policy. Produced by the PRE Dissemination Center Domestic Purchase Requirements for Import License Allocations in Mali by Wendy E. Takacs Table of Contents I. Introduction 1 II. Linking Arrangements with Competitive Industries 2 A. A Model of the Linking Arrdngement 2 B. Cost of the Linking Arrangement 5 C. Policy Objectives and Alternative Policy Instruments 6 D. Linki.ig Arrangements and Price Controls 9 III. Linking Arrangements with Monopoly 11 A. A Model of Linking Arrangements with Monopoly 11 B. Objectives and Policy Instruments with Monopoly 13 IV. Summary 15 References 17 Appendix 18 Endnotes 20 Figures 21 The original version of this paper was prepared as background material for the joint UNDP/World Bank Trade Expansion Program which provides technical and policy advice to countries wishing to reform .heir trade regime. The author would like to thank Jaime de Melo, Shanta Devarajan, Charles Pearson, and the participants in the International Trade and Finance Association's session on Trade Policy in Atlanta in December 1989 for helpful comments on previous drafts. I. INTRODUCTION As part of its policy of trade liberalization, the government of Mali decided to dismantle the state trading organization, SOMIEX, which had held a monopoly on the importation of milk, sugar, salt, canned tomatoes, tea, tobacco products, and fertilizers.' New procedures were adopted for the importation of these products.2 These procedures involve a two-stage process in which a prospective importer first must apply for an import authorization (autorisation d'importation), then, after negotiating with foreign suppliers, freight and insurance companies, and arranging for credit, may apply for a final import license (titre definitif d'importation). For two products, sugar and tea, importers are required to purchase a certain amount of domestically produced output before a final import license will be issued. Importers must purchase one ton of sugar within the country before they will be granted a license to import one ton of sugar. For tea the required ratio of domestic product purchased to import license granted is 0.75. Similar linking arrangements have been used by Niger and Malaysia for rice3, Iran for nylon and Colombia for hops. Such a system can also work informally. For example reports indicate that in Brazil deals are sometimes struck between a prospective importer and domestic producers in which domestic producers agree to withdraw their opposition to granting an import license if the prospective importer agrees to buy a specified amount from them. This note investigates the incentive system facing importers under a linking arrangement a-id their likely behavior. Section II analyzes the impact on imports, consumption and domestic production and the social costs under the assumption that all of the relevant "industries" (production of the product in 2 the domestic market, importing and distribution) are competitive, in the sense that there are a large number of firms involved and nothing deters new firms from entering the market. Section III focuses on the interaction of the linking arrangement with price controls, because the products in question also have been subject to this additional regu'lation. The overlapping regulations pose problems of timing in the liberalization process. Should price controls be phased out before, after, or at the same time as the linking a Langement? Analysis of the competitive market structure is useful to clarify the operation of the linking arrangement and set a baseline for comparison with noncompetitive market structures, but it may not be the relevant market structure in Mali. The complexity of the licensing system and the possibility of arbitrary decisions on import authorizations and licenses may deter entry into the "industry" importing and distributing the product. Moreover, for sugar, there is only one domestic refiner of sugar, which would effectively constitute a monopoly in refined sugar production. Section IV investigates the operation of the linking arrangements when domestic production is monopolized. The conclusion is that the ranking of the policy instruments as a means of obtaining certain objectives may switch if the domestic industry is not competitive. II. LINKING ARRANGEMENTS WITH COMPETITIVE INDUSTRIES A. A Model of the Linking Arrangement4 Assume that there are firms that act as distributors of a product within the domestic market. All output and imports are marketed through these firms, who both buy from the domestic industry and import. Assume that the country is small so that the world market price of the imported product (P,,) can be taken as given, that the imported product and the domestic good are identical, and 3 that both production and distribution are competitive industries. Suppose that importers must buy one unit of domestically produced output to receive a license to import "r" units. Thus r - M/Q, where M is the quantity imported and Q. is the amount produced domestically. Assume that quantity demanded is a decreasing function of the price to consumers (Pd) and that the quantity supplied by the domestic industry is an increasing functior. of price to suppliers (P.). These demand and supply functions can be expressed in inverse forms as: (1) Pd = D(Qd) (2) P, = S(Q,) Market equilibrium requires that domestic supply plus imports equal demand: (3) Q, + M = Qd Given the linking arrangement, M - ;Q,, so: (4) (l+r) Q. - Qd The distributors are assumed to maximize profits from the distribution of imported and domestic goods. Let lower case m and q, denote one distributor's imports and domestic purchases, respectively. Profits equal the differeT ce between the buyTng and selling prices, less distribution costs, which are assumed to be zero. Profits (II) will be given by: IT - (P4-PF,) m + (Pd-P,) q, Since for each distributor m - rq,, rl - (Pd-P.) rq, + (Pd-P.) q. Maximizing profit with respect to q, would 4.mply: (Pd-P.) r + (Pd-P.) - 0 or, rearranging, 4 (5) Pd - (l/(l+r)) P, + (r/(l+r)) P, Under the linking arrangement, distributors would be maximizing profits when the price to consutmers is a weighted average of world and domestic producer prices. Equilibrium prices and quantities under the linking arrangement would be determined by the market equilibrium cordition (4) and the equilibrium condition for distributors (5). Figure 1 depicts the equilibrium when r-l (as for sugar in Mali). S' shows the total amount available in the domestic market. It is equal to domestic industry supply (S), plus imports allowed under the linking arrangerent. Given that r-l, S' horizontally doubles S. D is the domestic consumer demand curve. Given (5) and r-l, equilibrium will occur at CL, where the price at which consumers would buy that quantity lies halfway between S' and Pw. The price to consumers wculd be Pd; the price paid domestic producers wo,uld be P.. At that price, the domestic industry would supply an amount QL. Distributors would be willing to buy an amount QL (-ab) from domestic suppliers at a price P., import an amount CL-QL (-bd) at the price P,, and sell the entire amount CL (-ad) at a price of Pd. An increase (decrease) in "r", the ratio of imports to domestic production, would rotate S' farther to the right (left), make the linking arrangement less (more) restrictive, and provide less (more) encouragement to domestic production. At first it may seem counterintuitive that distributors would buy from domestic producers at a higher price then they charge when they resell to domestic consumers, but they are willing to incur a loss on domestic purchases because those purchases allow them to qualify for import licenses and obtain the profits from selling imported goods. At the margin, the loss on an additional unit of domestic purchases (P.-Pd) exactly cancels the gain on an additional unit 5 of imports (Pd-P.). Also in total, the profjils from importing and selling at the higher domestic price (area ginl) are offset by the losses from selling at a price lower than that paid to domestic suppliers (area abge). B. Cost of the Linking Arrangement The linking arrangement causes distortions and imposes net costs upon society of a similar nature, but a different magnitude, as those from a tariff or other type of quantitative restriction. Referring again to Figure 1, if no purchase of domestic output were required to import, and there were no other trade restrictions or transportation costs, price would be Ps, and consume:s would purchase Cp. The increase in price to Pd causes consumers to cut back their purchases to CL and results in a loss to consumers equal to area eirj. Under free trade, producers would produce Q,. The linking arirangement increases producer price to P., increases output to QL, and results in a gain to producers equal to area abkj and a production efficiency loss equal to area blk. Area abge represents the loss to distributors who buy at P. and resell at Pd. Given that it is equal to the rents obtained by importing under license 'area ginl), which in turn is part of the consumer loss, distributors can be thought of as intermediaries who transfer area abfe of the consumer loss to domestic producers. The other part of the producer gain, area efkj, goes directly from consumers to producers. Similarly, the production efficiency loss (area blk) is a combination of direct loss to consumers (area fglk) and indirect loss to consumers, through distributors as intermediaries (area bgf). The linking arrangement has the s.me effect as a tariff rate (Pd-P.)/P. where the tariff revenue (-ginl) is used to subsidize the domestic industry. In this partial equilibrium framework the net cost of the regulation is the 6 sum of the production deadweight loss (area blk) and the consumption deadweight loss (area irn). Rough estimates of the magnitude of the costs of the linking arrangements for sugar in Mali indicate that in 1988 the annual cost of the linking arrangement was about 1,678 Fcfa per year, or about 5.75 million dollars. Appendix A explains the calculations underlying this estimate. C. Policy Objectives and Alternative Policy Instruments Bhagwati and Srinivasan (1969) have shown that the optimal policy instrument depends crucially on the policy objective. A linking arrangement could be used, at a net social cost, to achieve a variety of policy objectives: (1) increase domestic production above what it would be under free trade; (2) reduce consumption; (3) reduce imports; and (4) maintain a particular ratio of imports to domestic production. The relevant question is whether each goal could be achieved at a lower cost by using another policy instrument. This section compares the linking arrangement with alternative policy instruments such as tariffs, direct subsidies, and consumption taxes, to achieve alternative goals, such as maintaining a particular level of domestic production or consumption, or limiting imports to a particular level or ratio of imports to domestic production. The linking arrangement depicted in Figure 1 increased production from Qy to QL at a cost of area blk plus area irn. A tariff rate of (P.-P,)/P. would have the same effect on production, but would raise the price charged consumers to P,, which would cut back consumption to CT and create a consumption deadweight loss equal to area crm, clearly larger than the deadweight loss with the linking arrangement. The linking arrangement increases domestic p-oduction at a lower cost than a tariff, but a direct subsidy to the industry would achieve the same 7 production objective at yet lower cost. A subsidy of P.-P. per unit of output would shift the industry supply curve downward to pass through point 1, and the industry would supply QL at price Pv. The direct subsidy would avoid consumption losses but the revenue to finance the subsidy would have to be raised in some way, which may be difficult for a country with a fledgling fiscal system, or might impose distortions in other markets. If the goal is to encourage domestic production, and if a direct subsidy is practically or politically infeasible, the linking arrangement will achieve the goal at a lower cost than a tariff. If the objective is to reduce consumption to CL, a tariff rate of (P4-P,)/Pw would do so at a cost in efficiency losses of fsk plus irn. This is a Iower cost than that incurred under the linking arrangement becatuse the tariff does not provide as large an incentive to increase higher-cost domestic production. A consumption tax at the same rate alone would be the least costly policy, given that it would reduce consumption to C! while producers still face the world market price P. and thus maintain production at Q,. Countries may want to control imports for balance of payments purposes or to maintain self-sufficiency. This objective could be specified as a given level of imports or as a given ratio of imports to domestiz production. If the goal is to maintain a given import level a tariff wil' do so at a lower cost than a linking arrangement. Figure 2 sets the two policy instruments to result in the same import level (bd-st). A tariff rate of (P,-P,)/P. would result in a consumption deadweight loss equal to area tru, and a production deadweight loss equal to area svk. Compared to the linking arrangement, the consumption deadweight loss with the tariff is larger by the area tinu, but the production deadweight loss is smaller by area sblv. Since vl-un, and sv=tu, the cost of reducing imports to a particular level must be less with the tariff than with 8 the linking arrangement. In addition, the tariff provides revenue to the government while the linking arrangement transfers revenue to the domestic industry, which may be in private hands. If the goal is to maintain a given ratio of domestic p.oductior, to imports, then the linking arrangement is the less costly policy instrument. Figu-e 3 compares the linking arrangement (again with r-l) and the tariff rate (P,-P.)/P. that would result in the same ratio of domestic production to imports. Compared to the linking arrangement, the tariff leads to a smaller production deadweight loss (area cgfarea ekj). The extra cost of the tariff is equal to area deji less area ahgc, and area deji must be greater than area ahgc.' This result implies that a tariff is a more costly method of achieving a given ratio of domestic production to imports than an arrangement that links quota license allocations to purchases of domestic output. Given the relative rankings of policy instruments, what are the possible interpretations of why the linking arrangement is used in Mali, and what policy recommendations would follow? If the relevant industries were competitive, one possible interpretation is that government's policy objective is to maintain a given degree of import penetration, or ratio between domestic production and imports, and that the linking arrangement is the most efficient policy tool to achieve that objective. In this case the stanc I recommendation that the quantitative restriction (in this case the linking arrangement) be converted to a tariff would increase the costs of obtaining the objective. Another, more plausible, interpretation is that the government's objective is to increase domestic production, but the use of the first-best direct subsidy to the industry is not feasible because of the difficulties involved in raising 9 revenue or the distortions that would be imposed in other markets by the revenue raising activities. Under this interpretation the linking arrangement is the second-best policy to obtain its objective. In this case as well, conversion to a tariff would increase the cost of obtaining the government's goal. D. Linking arrangements and price controls The discussion in the previous sections assumed that market prices (both the price in the market between domestic producers and distributors and in the market between dis.ributors and consumers) were free to adjust to clear the market. For the products subject to the linking scheme in Mali (sugar and tea) this is not the case because both are also subject to p ice controls'. The existence of the overlapping price control regulations raises a number of issues concerning the net ef;ect of both sets of regulations and the effects of removing one s.et of re-.1ations but not the other. To in-.vestigate these issues, Figure 4 reproduces the market-determined equilibrium under the linking arrangement from figure 1 and shows the impact of adding price controls. The equilibrium prices and quantities with the linking arrangement are the by-now-familiar P., Pd, CL and Q;. Suppose that, in addition to the linking arrangement, price controls impose a ceiling of P' on the price at which distributors can sell to consumers. Giver, that this ceiling is below Pd, the price controls create a disequilibrium in the market between distributors and consumers. At the official price Pd consumers would be willing to buy Cc' at the controlled price but domestic producers would only be willing to sell Q2. Given the domestic purchase requirements to obtain an import license, only CC would be available in the market. There would be an excess demand of (CcL -CCL). 10 The two regulator- measures work at cross purposes. The price controls reduce the divergence between world and domestic selling prices and thus the gain from importing, so distributors have less incentive to purchase from domestic producers. Domestic output would fall to QL, anid the price paid suppliers would fall to Pc. Price controls dilute tne effect of the import licpnse linking arrangement on domestic production, and thus reduce the costs of the linking arrangement. One interpretation of why the authorities use both price controls and the linking arrangements for tea and sugar is that they are attempting to avoid the consequences of one set of restrictions by imposing yet another. The linking arrangement drives the price to consumers above the free-trade price, so if the policymakers want to stimulate domestic production but also have as a sepafate objective keeping consumer prices low, they are faced with a dilemma. Using both regulations may reflect an attempt to achieve both objectives, or achieve an administratively determined balance between them. Any attempt to liberalize and deregulate in the face of overlapping regulatory requirements must face the question of in which order to remove the restrictive measures. In the case discussed here, the removal of price controls while the linking arrangement remained in effect, would increase the price from P'd to Pd. If the intent is to liberalize the trac'e regime and reduce, eliminate, or phase out the protection to the domestic industry, this price rise would provide a false signal to the domestic industry as to the direction in which to adjust. Temporarily higher prices may encourage additional investment in an industry that will be reducing capacity as the protection is phased out. On the other hand, if the linking arrangement were abolished first, the price to the consumer would fall from PL to P,,, the price controls would be I1 redundant, and could be dismantled. Thus if one objective of the government is to keep consLumier prices low, a program of liberalization might better ach'ieve this objective if the linking arrangement were abolished before price controls were terminated. III. LINKING ARRANGEMENTS WITH MONOPOLY A. A Model of Linking Arrangements with Monopoly The discussion of linking arrangements so far has assumed that the industry being protected is made up of many small firms which individually have no influence over the market for the product as a whole within the importing country. In the case of sugar in Mali, there is only one domestic sugar refinery. Similarlv in other instances where these restrictions have been used, there was only one producer of nylon in Irar. and rice marketing in Malaysia ;s controlled by the goverEunielnt rice marketing board. This type of market structure raises the question of whether the conclusions above about the relative efficiency of the alternative policy instruments carry through if the structure of the domestic market for the affected good is a monopoly. To answer this question, suppose that there is only one domestic producer or supplier of the import-competing good. Given the incentive system facing distributors under the linking arrangernent, the monopoly domestic seller can exploit its market power to maximize profits. If the competitive distributors are maximizing profits, equation (5), above, showed their equilibrium condition, which can be rewritten as: (6) p. - Pd + r (Pd - Pw) Given that Pd is a function of the quantity demanded by (1), and P. can 12 be assumed to be given for a small country, (6) can be thought of as the inverse demand function facing the single domestic supplier under the linking arrangements. It shows the amount that distributors are willing to pay the domestic monopolist for various quantities of output, which is the amount for which they can sell the good on the domestic market (Pd) plus the value to them of the r units they can then import under the linking arrangements. In essence the linking arrangements increase the demand curve facing the domestic monopoly supplier. Figure 5 illustrates the impact of the arrangements on a domestic monopolist. The domestic demand curve for the product is D. If r-l, domestic producer output must equal one-half of consumer demand, so the price to the consumer and the domestic producer output must lie along line segment ij, which bisects the horizontal distance between the demand curve for the product and the vertical axis. The distributor's demand curve for domestic output is given by (6). If r-l, the price distributors are willing to pay for a given amount of the domestic producer's output can be found graphically by adding vertically (at each output level for the monopolist) the consumer price and the difference between Pd and Pu, at that output level. The result is shown by the dashed line DR in Figure 5. The domestic monopoly would maximize profits by equating marginal cost and marginal revenue, given this demand curve. The marginal revenue curve associated with the demand curve DR is the dashed line MRR. If marginal cost were given by the line MC, the monopoly would produce at QL and sell to distributors at the price P.. The price to consumers would be Pd, and total consumption of imports and domestically produced output would be CL. Given the linking arrangement, distributors buy from the domestic producer 13 at P., buy in the world market at P", and sell both domestically purchased and imported units to consumers at Pd. The distributors incur a loss of area abcPd on units bought from the domestic producer, but make equal profits of cdef on imported units, thus breaking even overall. As in the competitive case discussed above, the distributors act as intermediaries to transfer the wedge between domestic and import prices (which can be thought of as tariff-equivalent revenue) to the domestic producer. B. Objectives and Policy Instruments with Monopoly As in the competitive case, compared with free trade, the linking arrangement increases domes-ic output, reduces imports, and reduces consumption. In the competitive market structure, the linking arrangement was found to be more efficient than a tariff if the goal is to increase domestic output or to maintain a particular ratio of domestic production to imports. Do these conclusions follow through for the monopoly market structure? In the analysis so far, the country in which the monopolist produces has been assumed small, which implies that the world price is given at P.. A tariff would increase the price of imports by the amount of the tariff revenue per unit, but imports would be freely available at this price. The domestic producer could not charge a price higher than the world market price plus the tariff, so marginal revenue for the domestic firm would equal P,(l+t) at all output levels. If the objective is to increase domestic output to QL, a tariff would have to be high enough so that marginal revenue would equal marginal cost at point g, or, expressed differently, P.(l+t)-h in Figure 5. Domestic price would be lower with the tariff, which implies lower consumer, and therefore social costs of maintaining the chosen production level. This comparison implies that if 14 domestic production is monopolized, a tariff is a less costly method of promoting domestic production than the linking arrangement, however a direct subsidy to the producer ot gf per unit produced would still be the most efficient of the policies considered. If the objective is to maintaini a particular ratio of domestic production to imports, iii this case a 1:1 ractio, then the tariff would have to be large etiough to increase marginal revenue to the producer (Pw(l+t)) so that marginal revenue would equal marginal cost along line segment ij. Given the specific cost and demand conditions in Figure 5 a specific tariff of nm or an advalorem tariff of nm/P. would raise the monopolist's marginal revenue under the tariff regime so that it would equal marginal cost at point n. This maintains the chosen market share for the monopolist, but with higher levels of domestic production and lower consumer prices. Comparing the costs of the linking arrangement and the tariff, there is a tradeoff between the costs of encouraging greater relatively inefficient domestic output, and the gains from lower consumption distortions. The additional deadweight loss from the greater output would equal area gnmf, while the consumer gain would equal area drs. In the case shown in Figure 5, the extra production costs outweigh the consumer gains, so the linking arrangement remains less costly than the tariff. However, a lower and flatter marginal cost curve would imply a smaller tariff, resulting in larger consumer gains relative to the extra deadweight losses from relatively inefficient domestic output.' This implies that the existence of monopoly also may reverse the rankings of tariffs and linking arrangements as policy instruments when the objective is to achieve a given ratio of imports to domestic production. Linking arrangements become more costly when the protected industry is monopolized because the linking arrangement increases the degree of monopoly 15 power that can be exerted by a single seller. The economy suffers losses both from the exploitation of monopoly power and from the distortions inherent in the trade restriction. The comparison of tariffs and linking arrangements under competitive assumptions above indicated that the linking arrangement used by the government of Mali may be the optimal policy instrument, or at least the second best policy instrument, if the goal of the government is to maintain a given degree of import penetration or stimulate domestic production. Given that the market structure in the protected sugar industry can be considered a monopoly, the linking arrangement is less efficient than a tariff if the government's goal is to promote domestic production and may be less efficient if the government's goal is maintain a domestic market share. Given these conclusions, it would be wise for the government to consider elimination of these restrictions or conversion to tariffs. IV. SUMMARY As part of its program of trade liberalization, the government of Mali has dismantled the state entity that held import monopolies on a number of products, and has substituted an import licensing system. To obtain a license to import two of these products, sugar and tea, a prospective importer must purchase given amounts of domestic output. The volume of imports is thus "linked" to domestic output of the imported product. This paper investigated the economic impact of these linking arrangements under two types of domestic market structure, perfect competition and monopoly. Regulations that require importers to purchase from domestic sources before they can import have an effect equivalent to a tariff in which the tariff revenue is 16 transferred to the domestic industry as a subsidy. These regulations impose costs upon the countries imposing them, but the cost may be more or less than the cost of using tariffs, depending upon the objective of the policy and the structure of the protected industry. If the protected industry is a competitive industry, linking arrangements will be less costly than tariffs if the objective is to increase domestic production or maintain a certain degree of self- sufficiency, defined as a particular ratio of domestic production to imports. These rankings can reverse however, if the protected market is comprised of a single producer. Overlapping price controls and linking arrangements on the same products dilute the effectiveness of the linking arrangements and result in disequilibrium in the market between distributors and consumers. Removal of the price controls before the linking arrangement in a liberalization would drive up prices to both consumer and the producer, providing false signals and possibly increasing adjustment costs. 17 REFERENCES Bark, Taeho, and de Melo, Jaime, "Export Quota Allocations, Export Earnings, and Market Diversification", The World Bank Economic Review, Vol. 2, No. 3, 1988 pp. 341-348. Bhagwati, J. N., "On the Equivalence of Tariffs and Quotas" in R. E. Baldwin, et. al., eds., Trade. Growth, and the Balance of Pavents: Essays in Honor of Gottfried Haberler Chicago: University of Chicago Press, 1965. Bhagwati, J. N., and Srinivasan, T. N., "Optimal Inter7ent,.on to Achieve Non- economic Objectives", Review of Economic Studies, Vol. 36, No. 105, 1969, pp. 27-38. Corden, W. Max. The Theory of Protection Oxford: Oxford University Press, 1973. Jenkins, Glenn P. and Kwok-Kong, Andrew Lai, "Pragmatism and Stability: The Political Economy of Agricultural Pricing Policies in Malaysia, 1960-1983", mimeo, Harvard University, Harvard Institute for International Development, May 1, 1988. McCulloch, R. , and Johnson, H. G., "A Note on Proportionally Distributed Quotas", American Economic Review, Vol. 63, No. 4, 1973, pp. 726-732. Republique du Mali. Ministere des Finances et du Commerce. Direction Nationale des Affaires Economiques. "Note sur les Procedures D'importation du Sucre- Lait-The-Sel", mimeo, Bamako, 16 January, 1988. Republique du Mali. Ministere des Finances et du Commerce. Direction Nationale des Affaires Economiques. "Proposition de Strategie Pour la Mise en Oeuvre de; Mesures de Liberalisation des 4 Derniers Produits de Monopole de la Somiex", mimeo, no date. Stryker, J. D., J.-J. Dethier, I. Peprah, and D. Brean, "Incentive System and Economic Policy Reform in Mali" Associates for International Resources and Development, Inc., mimeo, June, 1987. Takacs, Wendy E. "An Analysis of Domestic Purchase Requirements for Import License Allocation" in Krosrow Fatemi and Catherine Mann, eds., Proceedings of the International Trade and Finance Association 1989, pp. 95-107. Takacs, Wendy E. "Mixing Regulations: A Mixture of Tariffs and Subsidies" in Krosrow Fatemi, ed. International Trade: Challenges for the New Decade New York: Taylor and Francis, (Forthcoming). 18 APPENDIX A This appendix presents some preliminary calculations of the cost of the regulations on sugar importation and marketing in Mali. In figure 3, the production cost of the linking arrangement was identified as the triangular area blk. This deadweight loss (Dp) is equal to: DP 1/2 ( P. P.) (QL, - Qf) - 1/2 (P. - P,) (dQ/dP) P/Q Q/P dP (Al) DP 1/2 (P. - PW) E, Q (dP/P) where Es is the domestic elasticity of s4pply of the product, and dP is the change in the price of the product to suppliers if it were to fall to the free trade level. The consumption cost of the linking arrangement was identified as area irn. This deadweight loss (D,) is equal to: DC 1/2 (Pd - P.) (Cf - CL) = 1/2 (Pd - P.) (dC/dP) P/C C/P dP (A2) D- 1/2 (Pd - PW) Ed C (dP/P) where Ed is the elasticity of demand for the product and dP is the change in the consumer price if it were to fall to the free trade level. From information currently available, the landed price of imported granular sugar in Bamako is 119,000 Fcfa per ton. Taxes of 38,681 Fcfa are imposed, so the selling price of imported sugar (P.) would be 157,681 Fcfa per ton if the taxes alone were applied. The official sales price of granular sugar (Pd) is 300,000 Fcfa per ton, and the factory gate price of domestically produced sugar (P.) is 254,780 Fcfa per ton. Sugar consumption in Mali (C) is estimated to be 39,000 tons per year. Given the linking scheme, domestic production (Q) would be one half of consumption, or 19,500 tons. 19 No estimated elasticities of demand for and supply of sugar in Mali are known to be available, so for illustrative purposes both are assumed to be equal to one. Given this data and applying equations Al and A2, the estimated production and consumption costs of the linking scheme and price controls is estimated to be: DP .5 (254,780-157,681.) (1) (19,500) (254,780-157,681)/254,780 .5 (97,099) (19,500) (.3811) - 361 million Fcfa D- .5 (300,000-157,681) (1) (39,000) (300,000-157,681)/300,000 .5 (142,319) (39,000) (.4744) = 1,317 million Fcfa The total deadweight 'sses, or cost of the linking .-rrangements and price controls is thus 1,678 milliLn Fcfa per year, or 5.75 million US dollars per year at 1988 exchange rates. These numbers must be taken as very rough estimates that illustrate the potential magnitude of the efficiency losses involved in the overlapping regulatory controls on sugar marketing. In the first place, they are based on a guess of the demand and supply elasticities. They also ignore distribution costs, which would increase the free trade price to the consumer and reduce the estimated costs. 20 ENDNOTES 1. See Republique du Mali, "Proposition..." for an explanation of the program. 2. See Republique du Mali. "Note..." for an explanation of the procedures. 3. See Jenkins and Kwok-Kong (1988), for a description of Malaysian rice policies and an analysis of the mixing regulation along the same lines as the approach taken here. 4. Corden (1973) presents an analysis of mixing regulations similar to that used here. Similar license allocation methods have been analyzed by McCulloch and Johnson (1973) and Bark and de Melo (1988). The model used here also appears in Takacs (1989) and Takacs (forthcoming). 5. Area deji - (Pt-PW)(CR-CT) - l/2(Pt-Pd)(CR-CT) Area abgc - (PI-Pw)(QJ-QT) + l/2(P.-Pt)(QX-QT) Let (CF_CT)_C'; (Q1-QT)-q' The extra cost of the tariff (X) is thus: X - (Pt-P)c-1/2(Pt-Pd)C -(Pt-Pv)q -1/2(P.-Pt)q' Because c'-2q' and (Pt-Pv)-(Pd-P")+(Pt-Pj) X - (Pd-Ps) q' - 1/2 (P.-P,) q' , which must be positive becau_e (Pd-P")-(P.-Pd), and (P.-P,) must be less than (P.-Pd). 6. See Stryker, Dethier, Peprah and Brean (1987) for a description of the price control system. 7. The reader can verify the reversal in the rankings of the instruments in Figure 5 by working through the case with a flat marginal cost curve that crosses MRR halfway between points g and k. 21 Figure 1 price S SI D QF QL CT CL CF quantity 22 Figure 2 price S a _ba SA I-m p~~~S iik/xl Pt_ D Q; Q L CT CL quantity 23 Figure 3 price S S I Ps C/ICT CL quant ity 24 Figure 4 )ri ce S PS Pd I I I I -II P dC I I I x PW C4L CL QL CL L quantity 25 Figure 5 price \ ' MRR DM QL CL quanti ty PRE Working Paper Series Contact Title Author LA for gaper WPS554 Korea's Labor Markets Under Dipak Mazumdar December 1990 M. Schreier Structural Adjustment 36432 WPS555 The Macroeconomics of Price Reform Simon Commander December 1990 0. del Cid in Socialist Countries: A Dynamic Fabrizio Coricelli 39050 Framework WPS556 Taxing Choices in Deficit Reduction John Baffes December 1990 A. Bhalla Anwar Shah 37699 WPS557 The New Fiscal Federalism in Brazil Anwar Shah December 1990 A. Bhalla 37699 WPS558 Alternative Instruments for Kenneth M. Kletzer December 1990 J. Carroll Smoothing the Consumption of David M. Newbery 33715 Primary Commodity Exporters Brian D. Wright WPS559 Fiscal Policy and Private Investment Ajay Chhibber December 1990 D. Bilkiss in Developing Countries: Recent Mansoor Dailami 33768 Evidence on Key Selected Issues WPS560 The Persistence of Job Security in Milan Vodopivec December 1990 CECSE Reforming Socialist Economies 37188 WPS561 The Labor Market and the Transition Milan Vodopivec December 1990 CECSE of Socialist Economies 37188 WPS562 Anticipated Real Exchange-Rate Luis Serven December 1990 S. Jonnakuty Changes and the Dynamics of 39076 Investment WPS563 Empirical Investment Equations in Martin Rama December 1990 E. Khine Developing Countries 39361 WPS564 Costs and Benefits of Agricultural Avishay Braverman December 1990 C. Spooner Price Stabilization in Brazil Ravi Kanbur 30464 Antonio Salazar P. Brandao Jeffrey Hammer Mauro de Rezende Lopes Alexandra Tan WPS565 Issues in Socialist Economy Stanley Fischer December 1990 CECSE Reform Alan GeIb 37188 WPS566 Measuring Outward Orientation in Lant Pritchett January 1991 K. Cabana Developing Countries: Can It Be Done? 37947 WPS567 Macroeconomic Management and the Antulio N. Bomfim January 1991 A. Bhalla Division of Powers in Brazil: Anwar Shah 37699 Perspectives for the Nineties PRE Working Pager Series Contact Ida Aulhor D for pale WPS568 Higher Wages for Relief Work Can Martin Ravallion January 1991 C. Spooner Make Many of the Poor Worse Off: Gaurav Datt 30464 Recent Evidence from Maharashtra's Shubham Chaudhuri 'Employment Guarantee Scheme' WPS569 Domestic Purchase Requirements for Wendy E. Takacs January 1991 S. Fallon Import License Allocations in Mali 37947 WPS570 Debt Concentration and Secondary Raquel Fernandez January 1991 S. King-Watson Market Prices Sule Ozler 31047