Article

A Supply Framework for Exchange Reform in Developing Countries: The Experience of Sudan

Author(s):
International Monetary Fund. Research Dept.
Published Date:
January 1980
Share
  • ShareShare
Show Summary Details

I. Exchange Rate Determination in Developing Countries

Exchange rate adjustment in developing countries with pegged exchange rates has frequently been approached from a macroeconomic framework with special focus on the traditional export sectors as one major cause of the disequilibria, but with overall emphasis on the corrective fiscal and monetary policies needed to bring domestic output and absorption into sustainable balance. The underlying framework of such an approach is one of the price or cost structure in a particular country moving out of line with the price structure of its trading partners by some cost/push or demand/pull phenomenon leading to a chronic disequilibrium in the balance of payments. Exchange reform is then justified by the purchasing power parity (PPP) theory and is brought about to realign the domestic price structure with that of its trading partners, with the support of some deflation and readjustment in income shares. And while the theoretical shortcomings of the PPP theory have been widely recognized, it frequently lingers as a modus operandi with policymakers.

The formulation of the exchange rate adjustment policy in terms of price and cost developments in the economy implicitly takes the current productive structure and growth path as given. Allocational shifts stimulated by price effects would induce a greater supply of exports as well as a restraint on import demand that would be satisfied eventually by an increase in the production of domestic substitutes for imports. The underlying assumption is one of relatively free market forces that would direct resources via price signals to the most productive uses while restraining demand for imports. Appropriate fiscal and monetary policies would then ensure that incomes would rise less than the overall output, thus freeing real resources to be directed to exports.

In a number of developing countries, most of these assumptions may not be valid. For one thing, large sectors of the economy may be administered by the public sector in pursuit of goals other than efficient resource allocation, and they may not be responsive to price and cost signals. Price fixing, domestic trade monopolies, acreage prescriptions for various crops, in addition to the usual decisions on taxation and investment, are only some of the policy instruments wielded by governments—often without any systematic attempt at consistency either among the use of instruments or among the variety of goals pursued. Consequently, price and cost distortions emerge, resulting in a cleavage between private incentives and social goals with policy-induced or institutional rigidities preventing the movement of resources in the desired direction.

Under conditions of protracted exchange controls and quantitative restrictions on trade, the structure of imports ceases to be responsive to changes in prices or incomes. Apart from basic food items, private imports for consumption may be reduced to a small proportion of total imports and may become unresponsive to price changes as a result of the scarcity stemming from the quantitative restrictions. The derived demand for the bulk of imports—raw materials, spare parts, and capital goods—would be virtually price inelastic as they emerge as a major constraint to the growth of output—particularly exportables. Reliance on demand management policies, in conjunction with the price effects of a devaluation, to reduce the demand for imports and to shift resources toward exports then becomes ineffective, unless the adjustment mechanism is also directed at providing the price incentives for export expansion and at removing the price distortions and supply bottlenecks that have hampered the growth of exports and have encouraged the production of inefficient import substitutes.

The balance of payments disequilibrium may also be a reflection of a skewed export structure, with a fairly sluggish demand abroad coupled with the pursuit of a growth rate and social goals that cannot be sustained by the earnings of this export structure. The attempt to move the economy to a higher sustainable growth path will then require that the basic parameters of the economy (which tend to be stable in a developed country), such as the rate of savings, the input/output structure, the ratio of trade to gross national product (GNP), be changed by the use of appropriate instruments. Hence, an exchange reform aimed at correcting such a balance of payments disequilibrium may also have to be an instrument of structural change, vested primarily with an allocative function toward the fulfillment of a production and trade structure that can sustain the desired growth rate.

In this context, the very notion of an equilibrium exchange rate takes a different dimension. Equilibrium is commonly defined as the rate that would result in overall payments equilibrium when all import restrictions and export subsidies are removed and full employment prevails.1 The time dimension implicit to the attainment of equilibrium is usually perceived as that which is necessary for the adjustment of the various price effects and output to be fully realized within the existing economic structure. The stringent requirement of moving an economy from a state of exchange controls and quantitative restrictions to a situation of free trade, apart from delivering a severe and possibly politically fatal shock to the system, has an unduly narrow time dimension of equilibrium. If a rate of exchange were predicated on the inducement of shifts in production and consumption, a balance of payments equilibrium might not result before such shifts fully materialize—a process that could take a number of years. Equilibrium could then be defined over a period of time that would allow for such shifts to occur and for the productive structure to acquire the capacity for earning foreign exchange that would sustain greater import liberalization.

To the extent that the exchange reform must be tailored specifically to the cost structure of the major export and import substitutes, to shift resources toward traded goods, and to foster structural change, the standard PPP diagnosis for such reform would tend to be inadequate in specifying the extent of the exchange adjustment required and in revealing its incidence on the profitability of the various traded goods sectors. Under its several variants, this approach focuses on price and cost-trend differentials between the home country and its trading partners. After choosing a base period when both the domestic economy and that of its trading partners are deemed to be in equilibrium, relative price differentials are derived by means of some index to indicate the extent to which the cost or price structure of the country has fallen out of line with that of its partners. This approach is sometimes modified by adjusting the price movements in keeping with changes in the effective exchange rate.

Given the highly diversified import and export baskets of industrialized countries, where exports from the home country compete not only with the other imports of its trading partners but also with their own exports, this approach may be an effective shortcut to measuring the overall competitiveness of an economy. This is particularly so when resources move freely, markets are unrestrained, and there are no sudden changes in productivity that may affect its cost structure. Even then, the approach raises a number of conceptual problems as surveyed by Officer and that fall outside the scope of this paper. 2 It may be worthwhile, however, to point out some of these difficulties as well as practical problems that arise under some of the conditions characterizing developing countries. Suppose that the country experiencing disequilibrium has an export basket limited to a few primary or agricultural commodities and possibly a few standard manufactures. The country’s exports do not compete with its imports nor with the commodities produced by its major trading partners—all of which are industrialized countries. Whether the traditional exports are raw materials or agricultural commodities, the rates of inflation incurred by the trading partners are only incidentally relevant to the competitiveness of these commodities, since the partners produce them only as substitutes. Of greater relevance would have been a comparison between price trends in the home country and price trends in costs of production in countries that produce competing commodities and that generally appear among developing countries as very marginal trading partners. For Sudan—a major exporter of groundnuts—this would imply a comparison of its price trends with those of Senegal, Nigeria, and South Africa, to name only the major competitors for these exports. If one were to include substitute oilseeds, the United States and Brazil would be added to the list. The multiplicity of competitors for each export, and even more so for substitutes, raises serious doubts as to the empirical feasibility of this approach. Moreover, price indicators in these countries are often limited to cost of living and wholesale price indices. The former usually reflects the price distortions introduced into the economy, as they are heavily weighted with such subsidized items as food staples or non-traded goods (housing, education, and health services). Wholesale price indices often include a substantial share of imports and do not reflect the suppressed inflation that results from the imposition of quantitative restrictions and is manifested by the emergence of shortages. Finally, in a number of developing countries, GNP deflators are simply weighted averages of the consumer and wholesale price indices that are used to deflate GNP series at current prices.

However, even if these price indices reflected broad movements of prices and costs, their relationships with costs of production in specific export sectors would be quite tenuous. For one thing, the commodity weights of the standard price indices would tend to differ markedly from the structure of these export sectors. Moreover, cost trends in long-established export sectors with their well-defined supply linkages and their own labor markets would tend to vary significantly from the changes in costs reflected by wage indices, and neither price trends nor wage indices tell us anything about developments in productivity in these export sectors, which may vary significantly among countries. There may be sudden shifts in productivity by the introduction of new technology, such as the gains reaped under the Green Revolution, or transitional decline in productivity resulting from labor migration and from the substitution of one factor of production for another. Moreover, the absolute levels of profitability may vary significantly among countries, as demonstrated by a number of empirical studies on manufacturing, 3 and some exports with substantial quasi rents may remain competitive for a considerable period despite higher rates of inflation. At the other extreme, should the export sector face a downward-shifting demand curve because of the emergence of substitutes or a change in taste abroad, even low domestic rates of inflation relative to those of trading partners may call for a depreciation of the domestic currency to encourage alternative export sectors.

The PPP approach is more meaningful with respect to the production of import substitutes, since rates of inflation in major trading partners coupled with movements in their exchange rates do affect the costs of imports that are competing with domestic production. However, the limitation of the price indices as indicators of movements in costs remains, and in any case, the production of import substitutes is shielded from foreign competition by high tariff barriers and quantitative restrictions, which would ensure their profitability in the domestic market regardless of differences in rates of inflation between the home country and its trading partners.

Given the limitations that the PPP approach acquires under the conditions described earlier in determining an “appropriate” exchange rate and the need for the policymaker to address himself to the various price distortions and supply rigidities that hamper the movement of resources in the desired direction, this paper suggests a “supply” framework for exchange rate determination by drawing on Sudan’s experience in focusing on the structure of the economy whereby the exchange rate—in conjunction with a number of corrective measures applied to the price and cost distortions—becomes a major instrument of export promotion and structural change. It is a partial approach insofar as it is not an optimizing exercise derived from a general equilibrium model that takes into account both demand and supply relationships. Limitations of data preclude the measurement of these relationships in Sudan as in most developing countries, and even, where feasible, a policy-oriented approach can ill afford to leave out certain complexities and information about the economy so as to conform to the simplifications prescribed by the current state of the art in model building. It is also partial to the extent that it does not take into account capital flows and the income effects of the supportive policies to an exchange rate adjustment that are left outside the scope of this paper. Its objective is to establish a broad operational framework with readily measurable quantities to assist policymakers in appraising their current exchange, trade, and investment policies and in formulating alternatives that would better correspond to their expressed goals. The approach pursued here consists essentially in deriving the cost structure of the economy by focusing on the production costs of major productive activities, revealing its competitiveness vis-à-vis alternative exchange rates, its areas of comparative advantage, and the trade possibilities that emerge therefrom, as well as the supply constraints and the demand limitations that have been manifested.

After an exposition of Sudan’s structural characteristics and imbalances in Section II, the main features of the approach pursued in this paper are discussed and Sudan’s competitive position in 1972/73 and 1976/77 is developed in Section III. Section IV presents an appraisal of the domestic policies pursued in the aftermath of the 1972 devaluation, and Section V gives the formulation of an exchange rate policy in the light of the 1976/77 results. The paper concludes with a discussion, in Section VI, of the supportive policies necessary to sustain the 1978 devaluation.

II. Sudan’s Structural Imbalance

The mainstay of Sudan’s economy is its agricultural sector. It is the source of virtually all exports and domestic food consumption, and provides inputs for a large proportion of industrial activity. 4 Structurally, the agricultural sector is composed of two distinct parts. The first is the irrigated areas concentrated mostly in publicly owned and administered schemes along the banks of the Nile River and its tributaries. The second is the rainfed areas, which for the most part are farmed by traditional methods but encompass a mechanized farm sector that is of growing importance. The irrigated agricultural schemes comprise large consolidated areas, such as the Gezira, which alone covers some two million feddans, 5 and a number of smaller schemes along the Blue Nile and the White Nile relying on gravity, flood, and pump irrigation. Rotation requirements, the necessity to organize water distribution, and the economies associated with the consolidation of small plots under one crop into large areas led to the establishment of central management boards that undertake the major production decisions, prescribe the cropping pattern, and provide the necessary inputs.

Cooperation between the boards, the farmers, and the Central Government involves a complex system of relationships, particularly with regard to the sharing of benefits. Historically, the Gezira and other public irrigation schemes were conceived for cultivating cotton as the only cash crop. Two other crops were cultivated for subsistence purposes, either complementing cotton in the rotation (a fodder crop) or using marginal land for the cultivation of sorghum—the basic food staple. The Government had a profit-sharing arrangement with the tenants whereby in return for land, irrigation, various services, and management, the Government and the Production Board would share roughly half of the net cotton proceeds, leaving the tenant the other half in addition to the full benefits of the subsistence crop. Incentives for tenants were concentrated on cotton cultivation as their sole source of cash and in line with the Government’s objective of promoting the cultivation of its major export crop. Yet, when the other cash crops—wheat and groundnuts—were introduced, the Government and the Board did not include them in their profit-sharing arrangements, leaving their profits entirely to the tenant. Other distortions in the tax system further deflated the tenant’s revenues from cotton, shifting his incentives to other crops. (See Chart 2.) 6 Since the acreage distribution among crops is prescribed to the tenant by the Board, his incentives are manifested in his crop husbandry and allocation of inputs among crops, which is ultimately reflected in his crop yields.

Chart 1.Sudan: Cropping Pattern in the Gezira1 Scheme, 1968/69-1978/79

(In thousands of feddans)

1 Including the Managil extension and Guneid.

Chart 2.Sudan: Distribution of Costs and Surpluses for Major Crops in Gezira, in World and Domestic Prices1

(In Sudanese pounds per feddan)

1 Based on costs for 1976/77 (Table 4). Tariffs, taxes, and subsidies obtained from the Ministry of Finance.

Sudan’s exports have been concentrated on a few commodities, with cotton accounting for about 63 per cent of the total, oilseeds 23 per cent, and the rest divided mostly between gum arabic and livestock products. The country is, therefore, extremely vulnerable to commodity-price fluctuations and adverse demand trends, particularly since Sudan commands, with Egypt, the position of a virtual duopolist in the world market for long-staple cotton. Therefore, there was a strong prima facie case for the diversification of Sudan’s export basket and possibly for the expansion of import substitutes. Thus, oilseed production was expanded in the irrigated areas, and a number of sugar projects were launched. But the largest expansion occurred in wheat and to a lesser extent in rice cultivation to promote “self-sufficiency,” following a sharp rise in the price of cereals in 1974/75. (See Chart 1.) Consequently, cotton acreage was reduced by 20 per cent to 30 per cent with the expectation that average yields would rise, since marginal cotton lands were given up for cereal cultivation. Yet, average yields actually declined, thereby compounding the shortfall in output and pointing strongly to the emergence of distortions in the price and cost structure, resulting in a divergence between private incentives and social benefits.

The impact of these policies on Sudan’s export performance was grave. A quantum index for exports shows a fall of 21 per cent in the volume of cotton exports (Table 1). The decline in exports between 1970 and 1977 did not result from the secular downward trend in demand for the long-staple variety of cotton. In fact, prices for Sudanese long-staple cotton rose rapidly over this period, partly because of the halving of cotton exports from Egypt and partly because of an overall strengthening of the cotton market. This can be seen from the export price index, which increased by more than 15 per cent a year during this period. The sharp increase in cotton prices afforded Sudan the opportunity of a substantial expansion in its cotton exports, rather than a retrenchment. Alternatively, had weakness in the long-staple cotton market been perceived, Sudan could have shifted its long-staple cotton acreage to a medium-staple variety, where, as a price taker, it could have expanded its exports considerably at the going price. Nor was the decline in exports the result of a surge in the domestic consumption of its exportables following the inflationary policies pursued during this period. Only a small proportion of Sudanese cotton production was being consumed domestically. Rather, the decline in cotton exports resulted from the government decision to diversify the productive structure of the economy, and the various distortions in the incentive pattern that this entailed. Although other export crops were expanded, and import substitution in wheat was pursued, they did not significantly offset the decline in cotton exports. Consequently, Sudan’s overall export performance, in volume terms, constituted a primary cause for the weakening of the balance of payments position during the 1970s.

Table 1.Sudan: Export Indices, 1970-78(1970-71 = 100)
197019711972197319741975197619771978
Volume index96104961014970908771
Of which, cotton 19410694882956707056
Value index 299101118138221193192243229
Export price index of industrial countries 397103112133163183185203234
Sources: Sudan export indices (Laspeyres) constructed from Bank of Sudan, Department of Statistics, Foreign Trade Statistical Digest. Export price index of industrial countries obtained from International Bank for Reconstruction and Development, Commodity Trade and Price Trends (IBRD EC-155).

Cotton’s weight in total exports is 62.5 per cent.

Adjusted to exclude the effects of the depreciation on unit values during the second half of 1978.

C.i.f. index of U. S. dollar prices of manufactured exports, to all destinations. Imports of manufactures from developing countries account for three fourths of Sudan’s total imports.

Sources: Sudan export indices (Laspeyres) constructed from Bank of Sudan, Department of Statistics, Foreign Trade Statistical Digest. Export price index of industrial countries obtained from International Bank for Reconstruction and Development, Commodity Trade and Price Trends (IBRD EC-155).

Cotton’s weight in total exports is 62.5 per cent.

Adjusted to exclude the effects of the depreciation on unit values during the second half of 1978.

C.i.f. index of U. S. dollar prices of manufactured exports, to all destinations. Imports of manufactures from developing countries account for three fourths of Sudan’s total imports.

While import price indices for Sudan are not as yet available, one can look at the index of manufactures in industrial countries, the source of most of Sudan’s imports with the notable exception of petroleum products. 7 This series, as a rough reflection of Sudan’s import prices did not outpace the rise in export prices and therefore does not point to a deterioration in Sudan’s terms of trade as a factor in the country’s balance of payments problem. In value terms, while exports grew at an average annual rate of 8.1 per cent between 1970-72 and 1977-79, imports increased more than two and one-half times as rapidly—at the rate of 21.0 per cent a year (Table 2). Most of the expansion in imports was in food, raw materials, and capital goods, while imports of consumption items other than basic necessities declined as a proportion of total imports, falling from 14.1 per cent in 1970 to 9.8 per cent in 1978. 8 Thus, while overall imports maintained their share in GDP at about 15 per cent, exports as a proportion of GDP fell from 14 per cent in 1970/71 to 7 percent in 1977/78. These trends give particular cause for concern, considering that under a normal development process, trade shares as a proportion of GDP would typically increase. 9 Such a process would have been expected for Sudan where monetization of the economy increased rapidly during the 1970s, and where only a decade ago, infrastructure was at a rudimentary stage.

Table 2.Sudan: Trade, Development Expenditure, Public Sector Deficits, and Credit Expansion, 1970/71-1978/79 1(In millions of Sudanese pounds)
1970/711971/721972/731973/741974/751975/761976/771977/781978/79Annual Rate of Growth
Exports109.2107.3127.490.3149.7191.7206.8191.9181.28.1
Imports109.9116.9120.9171.8322.3369.7343.3413.7460.021.0
Balance of trade-0.7-9.66.5-81.5-172.6-178.0-136.5-221.8-278.8
Net services-9.3-11.9-16.0-23.7-34.9-14.4-18.8-25.6-24.4
Current account-10.0-21.5-9.5-105.2-207.5-192.4-155.3-247.4-303.2
Development expenditure26.629.829.650.7102.4117.2159.4185.0198.0
Government deficit-22.2-25.3-29.3-54.5-128.7-143.5-199.2-130.0-245.0
Net domestic assets (end of period)152.2180.4206.1301.1399.7562.5721.7887.11,093.732.5
Sources: Bank of Sudan, Bulletin, and Ministry of Finance.

Fiscal years ended June 30; all trade values are based on customs records and are converted at the past official exchange rate of LSd 1 = US$2.87. Growth rates for trade are evaluated for the averages of 1970/71-1971/72 and of 1977/78-1978/79. Net domestic assets are defined as credit extended to the Government, public corporations, and the private sector minus “Other Items (Net)” of the banking system.

Sources: Bank of Sudan, Bulletin, and Ministry of Finance.

Fiscal years ended June 30; all trade values are based on customs records and are converted at the past official exchange rate of LSd 1 = US$2.87. Growth rates for trade are evaluated for the averages of 1970/71-1971/72 and of 1977/78-1978/79. Net domestic assets are defined as credit extended to the Government, public corporations, and the private sector minus “Other Items (Net)” of the banking system.

While a primary cause of the steady deterioration in Sudan’s balance of payments during the 1970s was the weak export performance, the steady buildup of inflationary pressures during this period also contributed substantially to the external imbalance. Much of the inflationary pressures generated can be attributed to the pursuit of expansionary fiscal policies by the public sector. The public sector in Sudan holds a large share in productive activities. In addition to its usual role of providing the economic infrastructure and utilities, it manages all the modern agricultural schemes in the irrigated areas, owns the bulk of industrial enterprises, and—through its various trading companies—monopolizes internal and/or external trade in certain basic commodities, such as wheat, cotton, oilseeds, and petroleum products. The expansionary monetary and budgetary trends can be readily seen from the aggregates shown in Table 2. While there was a strong case for a buildup of the country’s capital stock after a long period of stagnation of investment, too many project commitments were made within a short time, overextending both the administrative and physical capacities of the country and leading to high costs in execution and long delays in completion. This development expenditure has been supported mostly by foreign borrowing, as the generation of domestic savings remained low and current expenditure grew virtually as rapidly as government revenues. The growing government deficits were financed increasingly by resort to domestic credit and have been the major factor behind the rapid expansion in the net domestic assests of the banking system.

To remedy the situation while the external imbalance was still manageable, a devaluation of 15 per cent was undertaken in March 1972 in conjunction with stabilization measures. This, coupled with the Government’s decision to follow the U.S. dollar after its devaluation in December 1971 and the realignment of currency values in February 1973, resulted in an effective devaluation vis-à-vis most of its trading partners in excess of 20 percent. However, as a result of a number of factors discussed later, the steady deterioration in the balance of payments continued and necessitated a second (25 per cent) devaluation in June 1978.10

While this paper does not examine in an exhaustive manner the causes of Sudan’s weakening balance of payments position and the impact of the policies pursued during the 1970s, it is intended to present a systematic analysis of the competitiveness of Sudan’s export sector and major agricultural import substitutes and their responses to changes in various policies. A critical shift in policy was the decision in 1974 to alter the cropping pattern and the shift of incentives away from cotton; this raises several questions. Can the decision be justified on the grounds of efficient import substitution whereby savings on imports more than offset earnings forgone on exports? Were the tenants’ incentives consistent with the newly established cropping pattern? Should the shift in the cropping pattern prove to be detrimental to the maximization of net foreign exchange earnings, would not a correction of the incentive distortion, coupled with a reallocation of acreage among crops and a stabilization in aggregate demand, remedy the external imbalance? Or, to put it differently, were the two devaluations necessary? The answer to these questions lies primarily in an evaluation of the competitiveness of the Sudanese economy and its evolution over the period considered here. The questions cannot be answered satisfactorily by relating various price indices or by restricting the analysis to the movements of economic aggregates. Rather, they require a detailed analysis of the potentialities and the constraints affecting Sudan’s major productive activities, and the overall allocation of resources.

III. Competitiveness of Sudan’s Major Crops

The narrowness of the traded goods basket of developing countries, which considerably reduces the relevance of the PPP approach in evaluating their exchange rates, can be turned to an advantage by focusing directly on the costs of production of those goods and relating them to world prices. The latter, converted by the exchange rate into domestic c.i.f. import prices and f.o.b. export prices, can serve as benchmarks to determine whether domestic traded goods are competitive. Competitiveness would be evaluated along the lines initially developed by Bruno in his measurement of domestic resource costs and the various refinements that subsequently have been introduced by a number of authors 11 The starting point of such an evaluation is to derive actual costs of production by adjusting domestic costs and import prices for taxes, subsidies, and various other price distortions that emerge from administering prices and the exercise of various forms of monopoly power. Once real costs of production are estimated, one can measure the net foreign exchange earned on an export or saved on an import substitute per unit of domestic resources (nontraded goods and factors of production) used up in producing a given commodity. In effect, for each product, an implicit exchange rate is derived for the foreign exchange obtained (or saved) per unit of domestic resources used. This allows for a ready comparison of the various products in terms of their net foreign exchange returns as well as a comparison of these returns to the current exchange rates.

A proper measurement of the competitiveness of an economy should cover the entire spectrum of productive activities, encompassing the marginal activities as well as traditional exports and import substitutes. An exchange rate adjustment would need to ensure that the traditional exports are competitive and should be set at the margin of profitability of some of the import-substitute and marginal-export activities. The extent to which the new exchange rate can “capture” these activities would depend on their position on the scale of competitiveness and their expected supply response. Such an approach is possible for a country like Sudan whose productive structure is concentrated on a limited number of commodities with well-defined characteristics (e.g., agricultural commodities and standard manufactures) and readily available international prices. A more complex industrial structure, as in Korea or the Republic of China, with a diversified output structure that is oriented toward the consumer market would run into difficult problems of estimation because of product heterogeneity, joint products, and the multiplicity of inputs. However, a large number of developing countries have a relatively simple structure of production where this approach might be used fruitfully. 12

In addition to the “partial” characteristics mentioned in Section I, two limitations to this approach are worth pointing out. By setting the exchange rate at the edge of the marginal-export and import-substitute activities, one determines only the maximum exchange rate necessary to ensure the profitability of the “efficient” traded activities. While this exchange rate may be consistent with the objective of structural change and a reallocation of resources to those sectors found most competitive, additional information on the supply response would be needed to assess the impact of this rate on trade flows and to determine whether a lower rate might still be necessary to achieve a given balance of payments objective. Here, one must distinguish between the supply response of producers and the translation of this supply response into exports as discussed in Section V. Moreover, insofar as the desired supply response is contingent on expanding the infrastructure and the export-servicing capabilities, supply-elasticity estimates based on historical relationships would be of little help in determining the necessary price signals. Consequently, the determination of an exchange rate that would ensure the profitability of major exports and import substitutes as well as “infant” activities with a potential for expansion might be the best that could be done under conditions of structural change. Equilibrium would then be approached in discrete stages as supply capabilities unfold.

The evaluation of competitiveness has normative implications on the allocation of resources. Policymakers should encourage an expansion in the most competitive activities and cause a retrenchment in the inefficient ones. The underlying assumption to the competitiveness coefficients based on average cost of production is one of constant costs. This should fairly approximate cropping changes in irrigated lands (but not expansion in the irrigated area) and possibly cropping expansion in rainfed lands. However, such implications of allocation must be approached cautiously, as there might be sharp changes in costs from one set of cost conditions to another. This is illustrated later for sugar and more generally in the differences between irrigated and rainfed agriculture.

The evaluation of competitiveness for Sudan encompasses ten productive activities, all of which are agricultural or, for sugar, involve the processing of an agricultural commodity. With minor exceptions this framework is fairly representative of Sudan’s structure of production. In the irrigated areas, long-staple cotton is the traditional export crop and sorghum, the traditional food staple. Medium-staple cotton, wheat, groundnuts, rice, and, to some extent, sugar are all fairly recent attempts at diversifying production. In the rainfed areas, groundnuts and sesame can be considered as the traditional export crops, although partial mechanization of the latter has made possible a significant expansion in its cultivated area. Sorghum cultivation also benefited considerably from mechanization and was upgraded from its initial status as a subsistence staple to a cash crop with a significant export potential.

Competitiveness has been measured for two specific years—1972/73 and 1976/77. The former year was chosen because it preceded the commodity-price boom that followed the quadrupling of oil prices at end-1973. At the same time, it reflects the changes in the cost structure brought out by the realignment of currencies in December 1971 and the devaluation of the Sudanese pound in March 1972. It also preceded major changes in the cropping pattern in Gezira in favor of wheat and groundnuts and came at the end of a long period of stable crop distribution, when the position of cotton was unchallenged and tenants’ incentives were directed toward its cultivation. The latter year was chosen partly because of the need to make use of the latest available data on production costs. By 1976/77, the commodity-price boom, which peaked in 1974/75, had subsided into a more stable pattern of world prices—a necessary condition for the derivation of medium-term world prices that would serve as a basis for future price expectations. The choice of 1976/77 also offers the advantage of preceding the devaluation that took place in June 1978 and hence of being subjected to the same exchange rate regime as in 1972/73. While the discussion is centered on 1976/77, since it provides a reasonably current evaluation of competitiveness, a presentation of the cost structure and competitiveness in 1972/73 offers a valuable perspective on cost and productivity developments in various sectors and explains the causes that changed the absolute and the relative competitiveness of various crops. Further methodological aspects of the evaluation of competitiveness and the data coverage are discussed in Appendix I.

the competitive position of agriculture in 1972/73

For 1972/73, data on costs of production have been obtained only for the six irrigated crops; the results are presented in Table 3. Of all the irrigated crops, and prior to any policy move at diversification, long-staple cotton emerges as the most competitive. With a net value added, measured at world prices of LSd 71 per feddan, long-staple cotton contributed US$3.25 in export earnings per Sudanese pound of domestic resources used in its cultivation. This well exceeded the then official exchange rate of LSd 1 = US$2.87, thereby contributing substantial excess profits to the tenants and confirming Sudan’s strong comparative advantage in cotton production. Medium-staple cotton is second with a competitiveness measured at US$2.75 per Sudanese pound, which is somewhat below the official rate of exchange but exceeds the effective rate of exchange of LSd 1 = US$2.50 at which all output values and input costs are measured. Nevertheless, because medium-staple cotton did not benefit from the export subsidy, it was being taxed implicitly by the prevailing exchange regime, creating a disincentive to its cultivation. The ranking of medium-staple cotton below the long-staple variety is due essentially to lower world prices for this length of staple without being offset significantly by higher yields or lower costs of production.

Table 3.Sudan: Cost Structure and Competitiveness of Major Agricultural Field Crops in Irrigated Lands, 1972/73(In Sudanese pounds, except where indicated otherwise)
Cotton
Long-stapleMedium-stapleGroundnutsWheatSorghumSugar
1.Basic parameters
Average yields (kilograms per feddan)16817037005094744,400
World prices255.941.085.736.029.770.5
2.Output value per feddan (at 1972/73 world prices)393.080.344.821.118.0356.7
3.Input costs per feddan76.675.156.336.829.5255.24
a.Traded inputs at world prices22.424.113.013.38.679.9
Fertilizers2.52.7-2.61.66.3
Pesticides4.24.6-0.8-14.25
Seed0.60.61.72.40.3-
Sacks2.52.72.01.50.810.6
Machinery usage64.14.52.63.53.628.2
Transport8.59.06.72.52.320.6
b.Nontraded inputs at domestic prices16.416.419.76.74.733.9
Trade margins (commissions and interest)76.05.813.22.21.99.3
Storage, clearing, and handling 77.47.64.582.01.815.0
Irrigation93.03.02.02.51.09.6
c.Factor remunerations37.834.623.616.816.2141.4
Labor15.21012.01011.22.36.249.4
Land and water 1110.010.05.07.04.0-
Management3.63.61.51.51.527.3
Administrative 122.52.51.51.51.022.0
Tenant 136.56.54.54.53.542.7
4.Joint products---2.0-5.0-4.0-
5.Excess profits (2 + 4 - 3)16.45.2-9.5-10.7-7.5101.5
6.International value added (2 - 3a)
a.In Sudanese pounds70.656.231.87.89.4276.8
b.In U. S. dollars at 1972/73 prices176.5140.579.519.523.5692.0
7.Domestic resource costs (3b + 3d - 4)54.251.041.318.516.9175.3
8.Competitiveness in U. S. dollars at 1972/73 prices (6b ± 7)3.252.751.921.051.393.95
Sources: Gezira Board, Oilseed Corporation, Sugar and Beverages Corporation, and Bank of Sudan, Department of Statistics, Foreign Trade Statistical Digest. Land and water charges were derived from Appendix II.

Five-year weighted-average yields were centered on 1972.

Converted at the official exchange rate, LSd 1 = US$2.87. Prices for cotton are per bale of lint (420 lbs.); all other prices are per ton.

Converted at the effective exchange rate, LSd 1 = US$2.50.

Input costs per ton of refined sugar.

Chemicals.

Includes depreciation costs.

Based on detailed cost breakdowns available for cotton and oilseeds; estimated for other crops.

Includes decortication.

Estimates of water delivery costs by Ministry of Irrigation including operational expenses, maintenance of irrigation facilities and of the sluice gates in the Sennar Dam, and operational expenses incurred by the Roseires Dam. All expenses scaled by the number of waterings for each crop.

Includes tenant’s “individual” account.

Estimated in Appendix II.

Administrative operating costs of the Agricultural Board prorated at 50 per cent for cotton and 50 per cent for the remaining crops; for sugar, administrative costs incurred at Khashm el Girba plant.

For Gezira crops, tenant compensation for management estimated from historical income levels and checked against urban earnings of relatively skilled workers. For sugar and rainfed crops, based on a “normal” return to capital set at 10 per cent.)

Sources: Gezira Board, Oilseed Corporation, Sugar and Beverages Corporation, and Bank of Sudan, Department of Statistics, Foreign Trade Statistical Digest. Land and water charges were derived from Appendix II.

Five-year weighted-average yields were centered on 1972.

Converted at the official exchange rate, LSd 1 = US$2.87. Prices for cotton are per bale of lint (420 lbs.); all other prices are per ton.

Converted at the effective exchange rate, LSd 1 = US$2.50.

Input costs per ton of refined sugar.

Chemicals.

Includes depreciation costs.

Based on detailed cost breakdowns available for cotton and oilseeds; estimated for other crops.

Includes decortication.

Estimates of water delivery costs by Ministry of Irrigation including operational expenses, maintenance of irrigation facilities and of the sluice gates in the Sennar Dam, and operational expenses incurred by the Roseires Dam. All expenses scaled by the number of waterings for each crop.

Includes tenant’s “individual” account.

Estimated in Appendix II.

Administrative operating costs of the Agricultural Board prorated at 50 per cent for cotton and 50 per cent for the remaining crops; for sugar, administrative costs incurred at Khashm el Girba plant.

For Gezira crops, tenant compensation for management estimated from historical income levels and checked against urban earnings of relatively skilled workers. For sugar and rainfed crops, based on a “normal” return to capital set at 10 per cent.)

Groundnut cultivation in Gezira with a competitiveness of US$1.92 per Sudanese pound was clearly unprofitable in terms of the effective exchange rate set in March 1972. To a large extent, this results from the relatively low yields prevailing at the time and from the large export-marketing costs, which exceed the material input costs of groundnut cultivation. Since the world prices of groundnuts are passed on to the tenant (minus transport and marketing margins), the noncompetitiveness of the crop resulting in negative excess profits of LSd 11 would have induced him to shift his labor and inputs to other crops and to neglect its cultivation. However, since he is not charged for the land and irrigation costs (LSd 7 per feddan) nor for the administrative costs (LSd 1.5 per feddan), he still derived a positive return of LSd 3 per feddan for his managerial function. Moreover, groundnut cultivation in irrigated areas cannot be evaluated independently from cotton cultivation, which it benefits by adding nitrogen to the soil prior to the planting of the cotton. 13 The complementarity between cotton and groundnut cultivation can be taken into account by measuring the competitiveness of the crop rotation of groundnuts with the two varieties of cotton. For long-staple cotton in combination with groundnuts, the competitiveness would be US$2.68 per Sudanese pound, while for medium-staple cotton it would fall to US$2.38 per Sudanese pound.

Cereal cultivation in irrigated lands in 1972/73 appears to be noncompetitive. Wheat was the most inefficient crop, yielding only about US$20 per feddan in value added at world prices, which is about one tenth of the average value added per feddan obtained from the cultivation of groundnuts and cotton taken in rotation. Despite the fact that wheat is protected by a high transport barrier, since it is treated as an import substitute, it saves only US$1.05 per Sudanese pound of domestic resources used. Moreover, wheat may have an adverse effect on cotton and groundnut cultivation, since it uses stored water during the winter, possibly depriving these two crops of adequate irrigation in the spring when the water level is low in the Blue Nile. The major reasons for its inefficiency were the low yields obtained (about half a ton per feddan) and, as a relatively large user of water, the high costs of water delivery and land. However, its cultivation was encouraged by the Gezira Board on the grounds that as a winter crop it utilized land that otherwise would have been left fallow and machinery that would have been partly idle.

(See Appendix III.) Its cultivation was also valued by the tenants, despite losses amounting to half of its revenues at world prices, because of the high domestic producer prices (exceeding import price c.i.f. Khartoum by about 40 per cent in 1972/73) and, on the cost side, exemptions of any charges on land, water delivery, or administrative costs. Sorghum also emerges as noncompetitive in irrigated lands even when treated as an import substitute. Since its prices in Sudan are determined by market forces, farmers are not subsidized on this account. Rather, they are subsidized indirectly through the omission of any land and water charges from their cost structure—a significant amount relative to sorghum’s low output value per feddan, salvaging its profitability and accounting for the strong support its cultivation has among the tenants. 14

Sugar production in 1972/73 at a world price equivalent to US$0.08 a pound, which prevailed at that time, appears to have been quite competitive when evaluated in terms of the cost structure of the more efficient of the two existing sugar factories in Sudan. 15 Its competitiveness, when treated as an import substitute, almost approaches US$4 per Sudanese pound, substantially exceeding the competitiveness of long-staple cotton. If treated as an export, its competitiveness would be about US$3.30 per Sudanese pound. However, world prices for sugar during that period were moving along a sharp upward trend. Moreover, if sugar competitiveness were to be compared with cotton, its average competitiveness must be derived taking into account the much higher cost in the operation of the second plant. In this case, its competitiveness drops to about US$3.20 per Sudanese pound as an import substitute and to US$2.22 per Sudanese pound as an export. Either way, these results were sufficiently positive to have justified (at the time) the decision to expand sugar production on the assumption that world sugar prices would keep pace with the rise in sugar production costs, particularly its imported inputs. In the event, that assumption did not hold; while world prices of sugar stabilized at about the same level after the commodity-price boom in 1974/75 subsided, input prices more than doubled, thereby drastically reducing Sudan’s sugar competitiveness.

Turning to the overall competitive position of irrigated agriculture in 1972/73, the average competitiveness of the six crops considered here, weighted by their relative shares of output value, is found to be US$2.68 per Sudanese pound, or somewhat between the official and effective rates of exchange (Table 5). The degree of weighted-average competitiveness for agriculture as a whole could have been higher if data for rainfed crops had been available, since these have proven to be consistently more competitive than irrigated crops. On the other hand, it has been shown that the competitiveness of medium-staple cotton in rotation with groundnuts (US$2.38 per LSd 1) was somewhat lower than the effective exchange rate. Since there was a convincing case for expanded cultivation of these two crops on grounds of relative comparative advantage and for the sake of attaining some diversification in the export basket, a depreciation of the Sudanese pound would have been necessary to ensure a profitable expansion of these crops. Thus, the 1972 devaluation was certainly justified on these grounds, although a larger devaluation would have been more effective in ensuring the competitiveness of irrigated groundnuts and in encouraging manufacturing activity, particularly with the presumption (discussed later) that manufacturing in Sudan tends to be less competitive than agriculture.

Table 4.Sudan: Cost Structure and Competitiveness of Major Agricultural Field Crops, 1976/77(In Sudanese pounds, except where indicated otherwise)
Irrigated Crops
CottonRainfed Crops
Long-stapleMedium-stapleGround-stapleWheatSorghumRiceSugarGround-nutsSesameSorghum
1.Basic parameters
Average yields (kilograms perfeddan) 15685137634875476153,299347135323
World prices 211093154725011710916920545
2.Output value per feddan (at 1975-77 world prices) 3150.2126.987.840.327.357.9413.647.231.816.7
3.Input costs per feddan142.1137.891.464.446.6112.6406.2 440.430.417.9
a.Traded inputs at world prices35.141.723.326.911.554.4135.812.510.910.7
Fertilizers9.49.68.03.521.014.4
Pesticides11.716.41.26.66.1 5
Seed0.80.82.73.80.58.12.30.50.3
Sacks3.02.85.02.31.52.311.23.31.51.8
Machinery usage 65.06.67.27.32.012.271.15.85.4
Transport5.24.58.44.34.04.233.06.93.13.2
b.Nontraded inputs at domestic prices31.431.433.510.56.416.450.013.58.94.4
Trade margins (commission and interest) 713.713.723.33.02.44.014.410.67.03.4
Storage, clearing, and handling 712.212.26.5 82.91.35.120.02.91.91.0
Irrigation 95.55.53.74.62.77.315.6
c.Factor remuneration75.664.734.627.028.741.8220.414.410.62.8
Labor37.6 1026.7 1012.63.011.215.887.39.16.90.1
Land and water 1118.018.010.012.08.014.00.30.30.3
Management
Administrative 125.05.02.02.02.02.045.95.03.42.4
Tenant 1315.015.010.010.07.510.087.2
4.Joint products--2.7-7.0-5.5-2.0--1.0--
5.Excess profits (2 + 4-3)8.1-10.9-0.9-17.1-13.8-52.77.47.81.4-1.2
6.International value added (2 - 3a)
a.In Sudanese pounds115.185.264.513.415.873.5277.734.720.96.0
b.In U.S. dollars at 1975-77 prices287.7213.0161.233.539.68.7694.286.752.215.0
7.Domestic resource costs (3b + 3d - 4)107.096.164.530.529.66.2270.426.919.57.2
8.Competitiveness in U.S. dollars at 1975-77 prices (6b ÷ 7)2.692.222.461.101.340.152.573.222.682.08
Sources: Ministry of Agriculture, Gezira Board, Mechanized Farms Corporation, Oilseed Corporation, Sugar and Beverage Corporation, and Bank of Sudan, Department of Statistics, Foreign Trade Statistical Digest.

Five-year weighted-average yields were centered on 1976.

Average prices for 1976-77. Prices for cotton are per bale of lint (420 lbs.); all other prices are per ton. All prices are converted at the official exchange rate, LSd I = US$2.87.

Converted at the effective exchange rate, LSd 1 = US$2.50. Ginning outturn for long-staple cotton: 109 lbs. of lint, 206 lbs. of seed; for medium-staple: 120 lbs. of lint, 195 lbs. of seed. Shelling outturn for irrigated groundnuts, 65 per cent; for rainfed groundnuts, 70 per cent. Outturn for rice, 70 per cent.

Input costs per ton of refined sugar.

Chemicals.

Includes depreciation costs.

Based on detailed cost breakdowns available for cotton and oilseeds; estimated for other crops.

Includes decortication.

Estimates of water delivery costs by Ministry of Irrigation including operational expenses, maintenance of irrigation facilities and of the sluice gates in the Sennar Dam, and operational expenses incurred by the Roseires Dam. All expenses scaled by the number of waterings for each crop.

Includes tenant’s “individual” account.

Estimated in Appendix II.

Administrative operating costs of the Agricultural Board prorated at 50 per cent for cotton and 50 per cent for the remaining crops; for sugar, administrative costs incurred at Khashm el Girba plant.

For Gezira crops, tenant compensation for management estimated from historical income levels and checked against urban earnings of relatively skilled workers. For sugar and rainfed crops, based on a “normal” return to capital set at 10 per cent.

Sources: Ministry of Agriculture, Gezira Board, Mechanized Farms Corporation, Oilseed Corporation, Sugar and Beverage Corporation, and Bank of Sudan, Department of Statistics, Foreign Trade Statistical Digest.

Five-year weighted-average yields were centered on 1976.

Average prices for 1976-77. Prices for cotton are per bale of lint (420 lbs.); all other prices are per ton. All prices are converted at the official exchange rate, LSd I = US$2.87.

Converted at the effective exchange rate, LSd 1 = US$2.50. Ginning outturn for long-staple cotton: 109 lbs. of lint, 206 lbs. of seed; for medium-staple: 120 lbs. of lint, 195 lbs. of seed. Shelling outturn for irrigated groundnuts, 65 per cent; for rainfed groundnuts, 70 per cent. Outturn for rice, 70 per cent.

Input costs per ton of refined sugar.

Chemicals.

Includes depreciation costs.

Based on detailed cost breakdowns available for cotton and oilseeds; estimated for other crops.

Includes decortication.

Estimates of water delivery costs by Ministry of Irrigation including operational expenses, maintenance of irrigation facilities and of the sluice gates in the Sennar Dam, and operational expenses incurred by the Roseires Dam. All expenses scaled by the number of waterings for each crop.

Includes tenant’s “individual” account.

Estimated in Appendix II.

Administrative operating costs of the Agricultural Board prorated at 50 per cent for cotton and 50 per cent for the remaining crops; for sugar, administrative costs incurred at Khashm el Girba plant.

For Gezira crops, tenant compensation for management estimated from historical income levels and checked against urban earnings of relatively skilled workers. For sugar and rainfed crops, based on a “normal” return to capital set at 10 per cent.

Table 5.SUDAN: Summary of Results of Competitiveness and Cost Trends, 1972/73-1976/77
Irrigated Crops
CottonRainfed Crops
Long-stapleMedium-stapleGround-stapleWheatSorghumSugarRiceGround-nutsSesameSorghumWeighted Average Competitiveness 2
Competitiveness in 1972/73 23.252.751.921.051.393.962.68
Increase in output prices 397127801006855
Increase in productivity 3-17-279-415-25
Increase in imported input costs 35773791023470
Increase in domestic costs 3978858517554
Competitiveness in 1976/77 22.692.222.461.101.342.580.153.222.682.082.44
Import coefficient 324.730.225.441.824.733.448.330.935.859.8
Labor intensity 3-437.030.224.720.240.142.922.934.933.914.0
Hypothetical competitiveness in the post-1978 devaluation 52.442.022.231.001.212.350.142.932.441.902.21
Sources: Tables 3 and 4.

Weighted by relative output value shares.

In U. S. dollars per Sudanese pound; evaluated at the effective exchange rate, LSd 1 = US$2.50.

In per cent.

Labor intensity is defined here to include labor plus tenant remuneration divided by total input costs.

In U. S. dollars per Sudanese pound; evaluated at the effective exchange rate, LSd 1 = US$2.00, for output and imported input values with the assumption that domestic resource costs have increased by 10 per cent.

Sources: Tables 3 and 4.

Weighted by relative output value shares.

In U. S. dollars per Sudanese pound; evaluated at the effective exchange rate, LSd 1 = US$2.50.

In per cent.

Labor intensity is defined here to include labor plus tenant remuneration divided by total input costs.

In U. S. dollars per Sudanese pound; evaluated at the effective exchange rate, LSd 1 = US$2.00, for output and imported input values with the assumption that domestic resource costs have increased by 10 per cent.

the competitive position in 1976/77

Competitiveness in 1976/77 is evaluated for ten crops for both irrigated and rainfed areas (Table 4). The results derived for both 1972/73 and 1976/77 are summarized in Table 5 together with the changes in the various cost components and the derivation of an initial position in the competitiveness of the various crops following the 1978 devaluation.

Long-staple cotton remained the most competitive irrigated crop at US$2.69 per Sudanese pound; however, it could not compete at the official rate of exchange at which its export proceeds were converted. The competitiveness of sugar production at US$2.57 per Sudanese pound in the more efficient of the two plants remained above the effective exchange rate, and sugar was able to compete effectively with imports at a landed cost in Khartoum of about US$0.14 a pound. However, at an implicit export price of US$0.10 a pound, its competitiveness fell well below the effective exchange rate (LSd 1 = US$2.50). Groundnuts were third in irrigated areas at LSd 1 = US$2.46, while medium-staple cotton emerged as significantly noncompetitive at that rate. Since the export proceeds of both medium-staple and long-staple cotton were converted at the official exchange rate, and, in addition, pay more than their share of land and water charges and administrative costs, their cultivation would not have taken place had the tenants been free to determine their own cropping pattern. Given the losses that the tenant is forced to incur on this crop, it is no wonder that he refers to cotton as “the Government’s” crop. Wheat and sorghum cultivation appear again as high-cost operations as shown by their “savings” (as import substitutes) of US$1.10 per Sudanese pound and US$1.34 per Sudanese pound, respectively.

Crops cultivated in rainfed areas appear to be much more competitive than irrigated crops because they do not incur the heavy irrigation and administrative costs, nor most of the fertilizing costs. The competitiveness of oilseeds substantially exceeded the effective exchange rate with earnings of US$3.22 per Sudanese pound for groundnuts and US$2.68 per Sudanese pound for sesame. The competitiveness of sesame has been declining rapidly, however, because of particularly pronounced falls in yields—by as much as 30 per cent between 1971/72 and 1976/77. Here, labor shortages at harvesting time appear as the dominating factor and point toward mechanization as the only prospect for revitalizing this crop in the future. Sorghum, a lower-value crop, was found to be much less competitive at US$2.08 per Sudanese pound. While it could not compete for exports at the effective rate of exchange, it remained an efficient import substitute, earning US$2.74 per Sudanese pound when transport and marketing costs for exports were excluded.

The competitive position of agriculture in Sudan deteriorated between 1972/73 and 1976/77 (Table 5). The weighted-average competitiveness for the six irrigated crops considered in 1972/73 declined from US$2.68 per Sudanese pound in that year to US$2.26 per Sudanese pound in 1976/77. The overall weighted-average competitiveness of all the crops in 1976/77 was US$2.44 per Sudanese pound, implying a substantial decline in the competitiveness of rainfed crops as well. Most crops incurred a decline in their competitiveness, mainly as a result of a fall in productivity (yields) coupled with a rapid increase in domestic costs of production and no significant gains in the output/input external terms of trade.

Long-staple cotton benefited from a virtual doubling of its world prices, partly as a result of a substantial decline in output in both Sudan and Egypt. Its imported input costs rose by much less than output prices, but this was more than offset by the 17 per cent decline in yields and a virtual doubling of its domestic costs. While lack of incentives was partly responsible for this decline in yields, competing demands for water from other crops and the deterioration in the water delivery systems also played a major role. The rate of increase in domestic costs is quite significant for cotton, since this is a labor-intensive crop that has undergone virtually no changes in its production techniques and factor proportions. Consequently, the increase in the costs of labor and nontraded goods at the same pace as the increase in cotton export prices is indicative of the inflationary pressures that have prevailed over this period and the rapid rise in labor costs.

The competitiveness of medium-staple cotton deteriorated from US$2.75 per Sudanese pound to US$2.22 per Sudanese pound as a result of a sharp fall in yield and despite a large increase in its output prices. It therefore ceased to be competitive at the rate of LSd 1 = US$2.50 and fell into third place in irrigated areas behind long-staple cotton and groundnuts, whose relative positions improved. Medium-staple cotton was affected adversely by a sharp rise in both domestic costs and imported input costs. As with long-staple cotton, its labor costs increased rapidly while its imported input costs bore the brunt of the rise in oil prices, since cotton is a heavy user of pesticides and fertilizers. In contrast, neither of these two inputs is used in the cultivation of groundnuts, whose overall costs of production rose only moderately. The substantial increase in the imported input costs of cultivating groundnuts resulted simply from an upward shift in the capital/labor coefficient with the introduction of mechanical harvesting. But overall efficiency in groundnut cultivation increased with both a slower rise in costs relative to other crops and an increase in productivity. Consequently, its competitive position improved, rising from US$1.92 per Sudanese pound in 1972/73 to US$2.46 per Sudanese pound in 1976/77 (marginally below the effective exchange rate). Sugar production suffered a sharp fall in competitiveness as a result of lower productivity and increases in production costs, which exceeded the rise in its world price.

The positions of wheat and sorghum- in irrigated areas remained virtually the same as in 1972/73, with a very low level of competitiveness. The increase in the world price of wheat was offset by a rise in its imported input costs as a result of the mechanization of its cultivation and the rise in the price of fertilizers. Since it is one of the most import-intensive field crops, its cultivation benefited least (together with sugar) from the 1972 devaluation. For irrigated sorghum, which is grown using traditional techniques with little mechanization, the rise in its domestic production costs outstripped the increases in its world price. This precluded an improvement in its competitiveness that would otherwise have resulted from the higher yields. Rice, grown on a limited scale in Gezira since 1974 in compliance with “self-sufficiency,” emerged as the most inefficient crop, primarily because of its low yields and high imported input costs. These are almost as high as its output value measured at world prices, verging on the possibility of a negative value added.

Generally, the evaluation of competitiveness for 1976/77 confirms the results obtained for 1972/73, as there was little change in the relative rankings of the crops considered. In the irrigated areas, Sudan preserved its strong comparative advantage in the cultivation of cotton and groundnuts. Nevertheless, the much stronger performance of groundnuts in rainfed than in irrigated areas strengthened the position of cotton and cast some doubt on the relative efficiency of cultivating groundnuts in irrigated areas. However, irrigated groundnuts have been as profitable as medium-staple cotton, and until a more efficient crop is found it will remain the second-best alternative to cotton. The cultivation of cereals on expensive irrigated land remained highly inefficient and resulted in substantial foreign exchange earnings forgone. Had cotton, instead of wheat and sorghum, been planted on half a million feddans in public entities, net foreign exchange earnings between 1975/76 and 1978/79 could have increased by about US$250 million. 16 (See Appendix III.) Sugar would not have been competitive as an export, since its export prices remained mostly below US$0.14 a pound (up to the end of 1979) and production costs must have increased. Nevertheless, it has been competitive as an import substitute, and it did save the country substantial amounts of foreign exchange during this period. In rainfed areas, Sudan has a strong comparative advantage in oil-seeds, while sorghum (the largest crop in Sudan, grown on more than six million feddans) has been able to compete only as an efficient import substitute.

IV. Domestic Policies in the Wake of the 1972 Devaluation

The results obtained for 1972/73 and later confirmed with the broader coverage of activities for 1976/77 have several allocational implications. In the irrigated areas, the cultivation of cotton and groundnuts should have been expanded to the limit afforded by the water capacity of the irrigation canals. Likewise, at the prices prevailing in 1972/73 sugar would have been a good prospect for expansion, given the cost structure of the more efficient sugar estate. Wheat and sorghum should both have been cut back—the former to the point where its water requirement did not infringe upon the water demands of groundnut sowing in the late spring and cotton cultivation in the winter, and the latter to the point where it used only marginal lands after groundnuts had reached their maximum attainable area. Actual developments in the cropping pattern, however, were entirely different. Cotton cultivation was cut back by 20-30 per cent while wheat acreage was trebled to a point where it exceeded cotton acreage, and, even though there was subsequently some retrenchment in its cultivation, its average in 1976/77 remained over two and one-half times greater than in 1972/73 (Chart 1). This expansion was undertaken in the name of self-sufficiency in food staples, even though Sudan produces a large surplus of sorghum, and was stimulated by the historically high prices attained in 1973/74 and 1975/76. Sugar production was also expanded considerably by the development of several sugar estates, but it was not until 1977/78—long after the peak sugar prices of 1974/75 had fallen to about US$0.08 a pound—that the first consignment of sugar was produced from the first newly completed factory at Sennar.

In the rainfed areas where crop distribution and expansion responds solely to market forces and supply constraints, the export-price stimulus provided by the 1972 devaluation coupled with the implicit depreciation of the Sudanese currency through its peg to the U.S. dollar had a considerable impact on acreage and output. Between 1971/72 and 1977/78, sorghum acreage expanded by 57 per cent, from 3.3 million feddans to 5.2 million feddans. Groundnut acreage showed the greatest relative expansion, from 1.3 million feddans to 2.2 million feddans. Only for sesame was the increase in acreage small (8 per cent), because of particular problems that limit a rapid expansion of this crop. 17

While the availability of data and technological shifts brought about by mechanization do not allow for a rigorous derivation of supply elasticities for the rainfed crops, the acreage response to higher prices and profits exhibited so far by private mechanized farms provides evidence of a substantial supply response in Sudan. This is made possible by the availability of large tracts of fertile but uncultivated lands in the rainfed areas, as well as opportunities for reclaiming land along Sudan’s rivers by expanding their irrigation networks. Additionally, since Sudan’s yields are generally low, much higher returns could be obtained from acreage presently under cultivation by relatively small increments of input.

The dramatic supply response to price incentives in rainfed areas was not matched by the provision of ancillary services necessary for a rapid marketing of relatively perishable crops and ultimately acted as a binding constraint on export growth. Railroad facilities, which are the core of the transportation network in Sudan (given the long distances and the bulkiness of Sudan’s exports), proved to be inadequate in meeting freight demand despite a number of rehabilitation programs undertaken over the past two decades. Indeed, the rail network may have suffered a deterioration in its capital stock relative to its state in the late 1960s as a result of poor maintenance and lack of spare parts. This was dramatically illustrated during the winter of 1977/78 when a record groundnut crop harvested in the west of the country could not be shipped to Port Sudan—in response to strong export demand—over a period of an entire year, with the result that more than 80,000 tons were left to spoil. Consequently, the groundnut crop fell in the subsequent year to the point where a ban on new export contracts was enacted in March 1979. Exports of sorghum have likewise been constrained to a maximum of 150,000 tons a year by inadequate transportation and lack of processing equipment. Large sorghum crops without significant outlet to foreign markets have driven domestic prices down, acting as a disincentive to further output expansion. Additional services—such as storage, grading, and packing facilities—are mostly lacking, and oilseed-crushing mills are often antiquated, resulting in lower-quality oils.

While farmers who work in irrigated schemes have no control over their cropping pattern, since it is prescribed to them by their production boards, they do face a set of material incentives determined by the relative returns that accrue to them from the various crops. These returns result from the domestic set of prices as well as the taxes and subsidies that they face, and both influence their overall allocation of resources among crops. Although it was shown that the prescribed cropping pattern took the wrong direction, this could have been offset to some extent by a set of incentives that would have strengthened the position of the efficient crops. 18

However, the opposite was true. The incentive pattern that emerged from the domestic cost and price distortions has been broadly consistent with the wrong cropping pattern. Cultivation of cotton has been heavily taxed by the discriminatory exchange rate applied to it, export taxes, and substantial customs duties imposed on its imported inputs. In addition, cotton has been bearing the full burden of costs incurred by the production boards as well as the water delivery costs, which can be imputed to the profit shares of the Government and the boards. Consequently, the producers received only about 70 per cent of the export value of cotton as is clearly demonstrated by Chart 2, which contrasts costs and tenant income for cotton, wheat, and groundnuts at domestic and world prices. 19 At domestic prices, the present profit shares of the Government and the Gezira Board amount to less than the land and water charges that should have been imposed on cotton. Yet, various layers of taxes make up the difference and siphon off the entire surplus, which otherwise would have accrued to the tenant. The effective government share of the surplus, which is the sum of the taxes levied plus the Government’s profit share, amounts to 31 per cent of the output value, while the tenant’s profit share is negligible (1.3 per cent). In contrast, the tenant derives a substantial income from wheat (24 per cent of its output value), which is subsidized by minimum delivery prices that are set by the Ministry of Agriculture substantially in excess of world prices. Tenants have also been able to extract an indirect subsidy from the production boards by incurring debts for the cultivation of wheat through the use of Board machinery and other inputs, and later evading repayment by marketing their crops to private traders. On the cost side, wheat was being subsidized by not charging any of the land, water delivery, or Board management costs. Had these price and cost distortions been corrected, tenants would have suffered substantial losses on the cultivation of wheat.

Groundnut cultivation also resulted in greater income to the farmer than did cotton, even though it was less profitable at world prices. Groundnut cultivators were benefiting from the effective exchange rate, and, with the exception of an export tax, they were in effect obtaining the export price for their output (minus transport and marketing costs). Additionally, their costs were subsidized by the lack of land and water charges or Board management costs. Consequently, at domestic prices, tenants’ incentives would favor primarily groundnuts followed by wheat, while they would have little, if any, incentives in cultivating cotton. Without these price and cost distortions, cotton would be the most profitable crop to the tenants, followed by groundnuts. At current yield levels, wheat cultivation could have been sustained only if the domestic price were raised to about LSd 110 per feddan.

With respect to the overall financial management, subsequent to the 1972 devaluation, stabilization of aggregate demand became progressively weaker. While the Government launched an ambitious development program, fueled by expectations of considerable foreign aid in the aftermath of the oil boom, it was unable to mobilize domestic resources either through its budgetary operations or through surpluses from its public enterprises. Nor were there meaningful attempts at channeling private savings into development investments. Both current and development expenditure increased rapidly after 1972, while revenues lagged, resulting in a growing overall deficit and increase in the government recourse to the banking system (Table 2). The public enterprise system as a whole was incurring yearly losses, which were also financed from the banking system, resulting in high rates of monetary expansion. These contributed to the sharp rise in domestic costs of production shown earlier and the erosion in the competitiveness of agriculture.

To summarize, while the 1972 devaluation was necessary to ensure the competitiveness of potential export crops, it had little effect on moving exports from their stagnation and correcting the structural imbalance in the economy. To a large extent, the price effects of the devaluation were undermined by a number of policies and distortions introduced into the economy:

(1) Acreage allocation in the irrigated areas did not move in favor of the two export crops that would be responsive to the price effects of a devaluation and would contribute most to a correction of the external imbalance. In fact, the opposite happened, as cotton production was actually cut back in favor of wheat, and the large sorghum acreage in irrigated areas limited groundnut expansion. Likewise, the distortions introduced into the tenants’ incentives through the use of dual exchange rates, the profit-sharing system in irrigated public schemes, and the taxation mechanism resulted in disincentives to cotton cultivation.

(2) While the price effects of the devaluation did contribute to substantial increases in acreage in rainfed areas and in output, lack of supportive transportation and marketing facilities created a disincentive to output expansion and hindered exportability, resulting in depressed domestic prices.

(3) Aggregate demand was expanded rapidly by the pursuit of expansionary policies, resulting in a much faster increase in domestic resource costs than in export prices.

V. Exchange Rate Determination

The deterioration in the competitive position of agriculture by 1976/77 once again raised the issue of the appropriateness of the exchange rate structure. To correct for this deterioration, a 25 per cent depreciation of the Sudanese pound was undertaken in June 1978. The analysis of the policies that accompanied the 1972 devaluation demonstrated the critical importance of enacting a number of supportive policies that would unlock supplies and lead to the desired export response. These policies, which may include substantial investments in ancillary services, must in turn be taken into account in the formulation of the necessary demand stabilization measures and also determine the adjustment period.

Two competitiveness distributions of ten activities in 1976/77 are shown in Table 5. The first distribution shows results of competitiveness prior to the 1978 devaluation. The second distribution assumes a certain “slippage” by the devaluation-induced rise in the prices of traded goods to the cost of domestic resources, resulting initially in some erosion of the competitiveness gained from the devaluation. Theoretically, the slippage would vary across sectors according to their relative demands for different labor skills, the extent of their labor organization, and, indirectly, through their demand for nontraded goods. In Sudan, since the activities considered here are mostly agricultural with unorganized rural labor, there are no empirical grounds to differentiate the impact of the devaluation on the domestic resource costs of various crops, other than a general presumption that the slippage would tend to be greater in the irrigated sector than in the rainfed areas. 20 Consequently, in response to the depreciation that was put into effect in 1978, the slippage was hypothetically set at 10 per cent across the board for all crops, which would reduce the competitiveness of all crops by the same percentage. Thus, a substantial degree of money illusion is expected to prevail, resulting in a relative decline in the real wage and in the prices of nontraded goods.

As a shift in domestic demand from traded to nontraded goods takes place and as the expansion in output of exportables materializes, prices of domestic resources will tend to rise. The extent of this increase depends on the elasticity of supply of the various categories of labor. Sudan does have a shortage of skilled labor, which has been drained away by the remunerative wages in the Gulf region following the oil boom. Additionally, seasonal shortages of unskilled labor have appeared, particularly in the harvesting of cotton and sesame, which has been increasingly reliant on the migration of seasonal workers from neighboring countries. Consequently, any expansion in demand for exportables or nontraded goods is bound to result in higher wages. Insofar as the export crops would be actively promoted in conjunction with their servicing sectors and infrastructure, demand management policies can have only a limited impact on their wage levels, short of jeopardizing the structural change pursued. In a second phase, substitution would take place from traded into the relatively cheaper nontraded inputs and from capital-intensive to labor-intensive techniques, resulting in different input coefficients from those shown in Table 4. Such substitution would mitigate the cost effect of the devaluation and would tend to strengthen the relative competitiveness of those products that offer the greatest opportunity for input substitution.

The ranking of traded goods according to their competitiveness can be visualized as a supply curve of exportables as a function of the exchange rate. Competitiveness and the exchange rate may be drawn along the vertical axis, while the value of exports offered would be drawn along the horizontal axis. To obtain a decrease in competitiveness along an ascending axis, one would plot the inverse of competitiveness (domestic resource costs) and express the exchange rate in units of the domestic currency per U. S. dollar (Chart 3). The shape of the curve itself would depend on the estimated or expected supply response to increases in the domestic prices of the traded goods that would result from successive depreciations of the domestic currency. The appropriate exchange rate level could then be determined at a point in the curve that indicates the maximum supply response that can be induced and beyond which diminishing returns and supply rigidities would emerge, at least in the medium run.

Chart 3.Sudan: Hypothetical Supply Curves of Exportables, with Domestic Resource Costs Shown in Units of the Sudanese Pound per U. S. Dollar1

1 Costs of crops plotted for 1976/77; sugar, irrigated sorghum, and wheat treated as import substitutes.

In practice, discrete intervals separate the competitiveness of various activities, and their distribution becomes in itself suggestive of the point at which the exchange rate line should be drawn. For Sudan, two clusters of activities emerge—one where competitiveness ranges between US$2.69 for long-staple cotton and US$2.08 for rainfed sorghum, and another for wheat and irrigated sorghum at about US$1.10. At the tail ends of the distribution, there are rainfed groundnuts at US$3.22 and rice at US$0.15. The interval between the two clusters is large enough to categorize wheat and irrigated sorghum as inefficient in the sense that there was no attainable exchange rate, that is, through incremental change in the current exchange rate, that would make these two crops competitive. Therefore, the exchange rate line can be drawn in the upper range of the first cluster, or at about LSd 1 = US$2.10. This would ensure the competitiveness of two major crops—medium-staple cotton and irrigated groundnuts—for which large expansion plans have already been implemented and that could not be sustained at the previous effective exchange rate (LSd 1 = US$2.50). It also strengthens the position of sugar, the expansion of which is certain to raise average capital costs. 21 And, finally, it affords rainfed sorghum, for which a large potential export market exists, the necessary stimulus for expansion.

Cereal production in irrigated areas may become competitive during short periods when its world prices are at high levels, as in 1974/75. Its cultivation could then be justified over a broad range of exchange rates. However, the determination of a cropping pattern cannot be made on transitory price phenomena but must be determined on historical averages that take the full commodity-price cycle into account. In any case, insofar as there is substitutability between sorghum and wheat, on the one hand, and cotton and groundnuts, on the other hand, and historical price trends move in the same direction, the question does not arise, since the opportunity cost of cultivating the latter (cotton and groundnuts) will remain substantially higher than the benefits derived from cultivating the former at any exchange rate. It is only in those areas where substitutability in production has been substantially reduced, say, by the availability of water or by land fertility, and at substantially higher levels of yield 22 that it may become profitable to cultivate cereals at a depreciated rate.

With the assumed increase in domestic resource costs following the June 1978 devaluation, an erosion of competitiveness along the lines shown by the hypothetical post devaluation distribution is unavoidable. To discount such an erosion, the depreciation of LSd 1 = US$2.00 was put into effect in June 1978 to maintain the profitability of the various export crops. However, should such an erosion actually occur, long-staple cotton would require a smaller implicit export tax than the present exchange rate structure implies, while medium-staple cotton would just attain competitiveness at the rate of LSd 1 = US$2.00 without any export tax. Rainfed sorghum would fall below this exchange rate at US$1.90; however, as sorghum exports are expected to remain a small proportion of total sorghum output and would stem from the most efficient producers whose competitiveness would be larger than the average, they could be sustained for some time at the rate LSd 1 = US$2.00.

An area of concern, however, pertains to the cotton textile industry, which is almost totally reliant on medium-staple cotton as its raw material and which could not be analyzed here for lack of data. Should the competitiveness of medium-staple cotton decline to the exchange rate LSd 1 = US$2.00 when account is taken of the increase in domestic costs of production, there would be no margin for any inefficiency in the production of textiles to be “absorbed” by the competitiveness of the raw material. Despite Sudan’s position as a major producer of cotton, several factors may cast doubt on the competitiveness of the textile industry as an import substitute:

(1) The protective barrier resulting from transport costs is very small for textiles.

(2) Sudan does not have abundant labor with low wage costs nor does it have the skills required to operate capital-intensive technology in textile manufacturing, and hence does not at present have a factor-based comparative advantage in producing cotton textiles.

(3) Sudanese medium-staple cotton may be too fine and expensive to serve as a raw material for basic cotton cloth, which is the major consumption item among textiles in Sudan.

These factors may partly explain the losses incurred by domestic textile mills despite their subsidization, for a number of years, through the purchase of domestic cotton lint. They may also explain the large inroads made into the domestic market by imported textiles from Asian countries despite their subjection to a depreciated exchange rate through their imports under own exchange and high tariff duties. More generally, since manufacturing in Sudan does not benefit from a “natural” advantage as does agriculture, or from favorable factor-cost conditions, there is a presumption that it would tend to be less competitive than agriculture.

The preceding considerations indicate that unless the rise in domestic costs of production is effectively restrained within the assumed increase, the rise in domestic resource costs would jeopardize the competitiveness of medium-staple cotton, sorghum, and possibly textiles. Such circumstances would reduce the effectiveness of the 1978 devaluation and would underscore the importance of pursuing strict demand management policies and containing inflationary pressures.

It should be stressed that while the suggested exchange rate adjustment ensures the competitiveness of the major crops in Sudan—a sine qua non for their expansion—it also provides all agricultural production with a greater incentive for output expansion through the intensification of cultivation in irrigated areas and the further possibility for acreage expansion (as well as intensificiation) in rainfed areas. However, by drawing the exchange rate line at the margin of competitiveness for sorghum and medium-staple cotton, one implicitly makes a judgment that a lower exchange rate would not result in a significantly greater supply response. In the absence of estimates of supply elasticities, 23 it is a judgment based not on the farmers’ responses to price signals, which in the rainfed areas have appeared to be quite elastic, but on the capability of the transportation, distribution, and servicing systems to deliver the increase in exportables to foreign markets. The limitations in this capability, discussed at length in the following section, suggest that an exchange rate adjustment under conditions of severe supply constraints must be approached cautiously and, if needed, in stages until the necessary infrastructure expands sufficiently to absorb increases in supply.

The determination of the exchange rate in terms of the cost structure of the economy’s major products also helps to assess the impact of the depreciation on the cost of their inputs and, by extension, on income distribution. Because the depreciation favors those commodities that are intensive in their use of domestic resources, it would further strengthen those crops that were found to be the most efficient (Table 5). Thus, rainfed areas would reap a relative advantage over irrigated areas. Rainfed groundnuts, which are cultivated by traditional methods and utilize hardly any imported inputs, would benefit most, followed by sesame, while sorghum cultivation, which has a large import content (57 per cent), would benefit least. In the irrigated areas, the position of groundnuts and long-staple cotton would improve relative to that of medium-staple cotton or sugar, while wheat, with the lowest domestic resource content (27 per cent), would benefit least. The depreciation would have a desirable income-distribution effect among producers, since it favors those commodities cultivated by the poorest producers—the tribesmen in the rainfed areas. Moreover, since the consumption basket of the lower-income groups in Sudan has a high proportion of non-traded goods and native food staples, while that of the higher-income groups is more import intensive, the burden of the adjustment would be relatively lighter on the lower-income groups. Nevertheless, even though there may be a relative improvement in the position of those groups, any reduction in their real income would represent a greater hardship than that sustained by the upper-income groups. This raises the often-neglected equity considerations in the adjustment process, a worthy subject for further research. Suffice it to say that historically, and particularly under present conditions characterized by labor scarcity, the subsistence level of the lower-income groups in Sudan, as measured by their intake of calories and protein as well as their access to housing and education, would compare favorably to the subsistence level in most of the neighboring countries. 24

VI. Supportive Policies

To break away from the export stagnation that has characterized Sudan’s economy over the past decade, it has been suggested that the exchange rate should be specifically tailored to the cost structure of the economy. The correction of price distortions and the reorientation of the cropping pattern toward exportables would not have been sufficient to promote exports, because a number of exports would have been unprofitable at the previous exchange rate. Conversely, given that large sectors of the Sudanese economy are administered by the public sector, changes in relative prices brought about by a devaluation would not in themselves result in the desired supply responses and reallocation of resources. A series of policy decisions that would be consistent with the objectives of the devaluation must be taken jointly to ensure that resources move in the desired direction and to preclude the policy inconsistencies that were followed subsequent to the 1972 devaluation.

Since the crop distribution in irrigated public schemes is prescribed to the tenant independently of relative prices and incentives, a reorientation of the cropping pattern toward the most competitive crops must be carried out jointly with the exchange rate adjustment. In the Gezira scheme, this would imply an expansion of cotton cultivation, the maintenance of present groundnut acreage, a considerable reduction in wheat acreage, and the phasing out of sorghum cultivation. The expansion of cotton cultivation in Gezira by 300,000 feddans coupled with a reduction in sorghum and wheat acreage would increase net foreign exchange earnings by US$55 million a year. (See Appendix III.) The reorientation of cropping must also be coordinated with a change in tenants’ incentives, so that their allocation of inputs among crops results in the highest attainable yields under the overall constraints of the availability of water and labor. Much progress has already been achieved in this respect. The domestic producer prices of exportable crops (cotton, groundnuts, and sorghum) in irrigated areas did increase by at least 25 per cent following the devaluation. In June 1979, the exchange rate applicable to cotton export proceeds was changed from LSd 1 = US$2.50 to LSd 1 = US$2.00, which placed cotton on a par with all other exports. At the same time, its export tax was abolished and the subsidization of cotton lint to domestic mills was ended. Land and water charges have also been imposed on wheat and groundnuts in the Gezira scheme. The profit-sharing formula on cotton has not been replaced by a flat land and water charge but has been amended to yield progressively larger profit shares to tenants with higher productivity. Despite these adjustments to the incentive pattern, charges on wheat and groundnuts are quite low relative to the imputed charges on cotton, and, in absolute terms, well below charges that would cover water delivery and utilization costs as well as provide the Government with an adequate return on its investment (Appendix II). Moreover, irrigated sorghum has been exempted from any charges that, when combined with the rapid rise in sorghum prices, would strengthen the tenants’ resistance to phasing out its cultivation. Nevertheless, the combination of price increases and the imposition of land and water charges will reorient tenants’ incentives in the desired direction and should result in higher yields for the exportable crops. Over time, the land and water charges will need to be adjusted upward and should be extended to other schemes and to all crops.

A third area of government policy in support of the devaluation pertains to the reordering of investment priorities. The decline in productivity during the 1970s, which undermined the 1972 devaluation, would have to be reversed if the current competitiveness of various crops is to be maintained in the face of rapidly rising costs. While this decline can be partly attributed to distortion in incentives and possible shortages of imported inputs, the deterioration of the infrastructure (e.g., water delivery, transportation), the lack of research and development of higher-yielding strains, and labor shortages also appear as major contributing factors. Raising yields and expanding the production of exportables will depend to a large extent on the timely provision of imported inputs, particularly fertilizers, pesticides, agricultural machinery, and packing material. Water delivery systems in New Halfa and the large number of pump schemes along the Blue and White Niles would have to be restored to their normal levels of operation. Growing labor shortages may need to be alleviated through the introduction of mechanization for harvesting medium-staple cotton and the expansion of mechanization for sorghum and oilseeds in the rainfed areas. Shortages of basic commodities, such as fuel, sugar, coffee, tea, and clothing, in rural areas owing to transportation difficulties or lack of foreign exchange may undermine the improved incentives that are being extended to the producers. One of the major reasons for the decline in gum arabic exports stems from the acute shortages of these commodities in the entire Savannah belt where tribesmen tap the gum trees. Consequently, the tribesmen have tended to flock to the populated areas, turning to less productive occupations but ensuring themselves of better access to consumer goods. Unless higher incomes can be translated into greater availability of basic consumer goods in the producing areas, the incentive for higher output will recede.

Increased production of exportables in response to better incentives and a reallocation of resources would be transformed into foreign exchange proceeds only if these exports are actually processed and transported to Port Sudan for shipment abroad. Lack of adequate transport and other export-servicing facilities in the past has acted as an effective constraint on export trade and has become one of the most potent arguments against the manipulation of relative prices, including the exchange rate, to increase trade flows. Unless development expenditure is concentrated on the rehabilitation and expansion of the transportation network, the increase in exports will not materialize. This is a major undertaking, given the deterioration that the railroads have suffered in their capital stock and the shortage of spare parts and fuel, which has reduced the capacity utilization of road transport. A diversification of exports toward oilseeds, cereals, and livestock has substantially increased the perishability and the bulkiness of the country’s exports, adding a further measure of urgency to the provision of adequate transport facilities. Likewise, export-servicing facilities, such as ginning plants, oilseed-crushing mills, decortication sheds, storage, and grading, would need to be developed to ensure the marketing of exports. Finally, adequate power must also be provided on a steady and reliable basis. Over the past few years, power cuts and fuel shortages have resulted in large losses of output.

The strengthening of Sudan’s infrastructure and the provision of export-servicing facilities to ensure shipment and marketing of exports abroad will require considerable investment and much larger operating costs. No estimate has been made for the increase in imports that would stem from a full implementation of the structural changes in the agricultural and transportation sectors, or from the alleviation of most of the current shortages. Nevertheless, such an increase may result in a quantum jump over past trends, unless there is a reallocation of foreign exchange resources from consumption items imported under own exchange to imports of production inputs. 25 A reduction in the launching of new development projects under the reordering of investment priorities will affect imports only after a considerable lag, while expanded production of exportables will require, ex ante, increases in imports. Since the reorientation of the cropping patterns toward exportables and the translation of a new incentive pattern into higher yields will take time to gain momentum, the supply of exports is not expected to increase significantly over the next two years. Moreover, an increase in export shipments is contingent upon the rehabilitation of the relevant infrastructure and the provision of the necessary imported inputs. This points to the conclusion that the viability of the exchange rate adjustments in Sudan will require large external financial assistance for balance of payments support. The adjustment period would have to accommodate the structural changes in the economy that are necessary for the full realization of the export potential.

It was pointed out earlier that the June 1978 devaluation would allow for an increase of only 10 per cent in domestic resource costs per unit of foreign exchange earned or saved before profitability begins to fall below the assumed level and acts as a disincentive on output expansion. 26 This would imply relatively stringent, but selective, demand management policies, particularly vis-à-vis urban consumption. A substantial increase in the generation of domestic savings would need to be actively pursued in support of the large investments envisaged in the restructuring of the economy and in view of the external debt burden. While there is some evidence of growing private investment, there has been little self-financing of investments by public sector firms. Whatever surplus has been generated by financially successful public enterprises, such as the Gezira Board, trading corporations, or banking institutions, has been more than offset by losses incurred by some agricultural and industrial enterprises. While the performance of these enterprises may improve with adequate supplies of raw materials and power, the challenge facing the Government is in reorganizing public sector enterprises to bring them to profitability and to generate a steady flow of public savings.

A full implementation of the export promoting measures outlined earlier is bound to increase the overall level of current government expenditure, particularly in areas governed by the Ministries of Agriculture, Irrigation, and Transportation. Yet government revenues, mostly from indirect taxes, have been found relatively inelastic with respect to income. 27 Thus, the burden of reducing the extent of deficit financing, which has fueled inflationary pressures in the past, must fall upon a broadening of the tax base and the generation of greater government revenues. A number of sources of potential government taxation have been left largely untapped, particularly in the rainfed agricultural estates and trade and construction sectors. In the long run, the generation of greater domestic savings becomes a necessary condition to sustain the development momentum of the past six years, which hitherto has been financed almost entirely from foreign savings.

The devaluation introduced in Sudan in June 1978 was carried out under difficult circumstances. In a situation of acute foreign exchange scarcity, which has acted as a binding constraint on output growth, a devaluation aimed at the expansion of the external sector would have to be backed by the availability of considerable foreign exchange to alleviate existing shortages and to generate a supply momentum in the economy. The infusion of imports, which such a foreign exchange cushion could finance, would also limit price increases to the direct effects of the devaluation, at least in the short run. However, in the absence of such foreign exchange or large external assistance, the price effects of the devaluation were compounded by widespread shortages throughout the production structure of the economy. This was further aggravated by the heavy rains that flooded large areas in the Gezira and other irrigated schemes. The losses caused by these floods contributed largely to a substantial decline in the production of Sudan’s major crops and to an expected overall decline in GDP. The export capacity of the country fell to its lowest level in a number of years in terms of both its primary factors (acreage distribution among crops and input avail-abilities, including labor) and the entire export delivery capability, particularly transportation. The phasing of the restructuring of tenants’ incentives by postponement of the imposition of land and water charges by a year, coupled with a gradual shift envisaged for the crop distribution in irrigated areas, will lengthen the adjustment period necessary for a major turnaround in the balance of payments, when exports begin to grow faster than imports. Nevertheless, much progress has been achieved in the past two years with the initiation of fundamental changes in demand management, agriculture, investment, and pricing. The ground has been laid for an economic expansion propelled by the export sector. It remains for domestic management and external financial assistance to bring it to fruition.

APPENDICES
I. Methodological Aspects of the Evaluation of Competitiveness for Sudan’s Major Crops

Since transport costs are considerable in Sudan and offer natural trade protection, a commodity may be competitive if treated as an import substitute and noncompetitive if treated as an export. In this paper, sugar, wheat, irrigated sorghum, and rice are all treated as import substitutes by adding to their c.i.f. import price the sizable cost of transportation from Port Sudan to the largest consuming center (i.e., Khartoum). In comparison, the domestic costs of production of these commodities incorporate only minor transport costs (e.g., from Gezira to Khartoum) and do not include all the marketing and handling expenses that exportability entails. However, since both sugar and sorghum have a major export potential, their competitiveness as exports has also been appraised.

Some notable omissions to the activities studied here are caused by lack of data. In agriculture, gum arabic is the most traditional of Sudan’s exports, accounting for 75 per cent of total exports at the turn of the century and reduced to 16 per cent in the 1970s. Its tapping is mostly a tribal activity, with labor as its sole input and virtually no information on its cost of production. Under its labor-intensive traditional structure, it is bound to decline further as an export activity because of strong competition from alternative occupations. Livestock breeding is another tribal activity for which little information is available. However, as with gum arabic, a significant expansion in its export is possible only under modernization of its techniques with a completely different cost structure. But such modernization is only in its incipient stages and would not enter as a factor in determining the exchange rate. Textiles and cement are major omissions in manufacturing. The former is a rapidly expanding industry, and its evaluation would have to await the full-capacity utilization of its newly erected integrated textile mills and the detailed cost records when they become available. Over the past few years, the two cement plants have been plagued by power cuts and low-capacity utilization, which preclude a meaningful derivation of the production costs for cement.

Since the evaluation of competitiveness in any single year is quite sensitive to the prevailing world prices and crop yields, it is important to derive representative world prices and yields that are not subject to erratic yearly disturbances. Five-year average yields were derived for 1972/73 and 1976/77, but for world prices, two-year averages were used at the tail ends of the time series to minimize the effects of the commodity-price boom in 1974/75. Adjustments to the data consisted mostly of netting out excise taxes, local taxes, and import duties from the input costs. Should an input be produced domestically at a higher cost than the c.i.f. price of the equivalent imported input, the latter would be used, sparing the output the penalty incurred in the utilization of an inefficient input. And while nontraded inputs are evaluated at domestic prices (minus taxes), in cases where such an input is a heavy user of an imported material (such as transport, the cost of which can be attributed mostly to the imported rolling stock and fuel), it is treated as a traded commodity evaluated at world prices. This provides a closer approximation of the direct and indirect import contents of an activity and minimizes the inclusion of import duties levied on the secondary inputs.

With respect to costs of factors of production, a set of land and water charges were imputed for all the irrigated crops in relation to the recurrent and capital replacement costs of the irrigation facilities and the watering requirements of each crop. (See Appendix II.) Because Sudan does not have a labor surplus and wages do respond to market forces, labor costs were taken at their market value. As for capital costs, a “normal” rate of return of 10 per cent was imputed to invested capital in mechanized farms and in the sugar industry. This is certainly low in relation to the rates of return expected by private entrepreneurs who venture their own capital and in relation to the rates of inflation that have prevailed in Sudan, particularly over the past three years. Nevertheless, it is above the actual costs of borrowing the capital invested in these projects, as it stems mostly from development project loans at low interest rates. Since Sudan will continue to rely heavily on foreign project loans at concessionary terms, for much of its investment in modern agricultural projects and industry, a 10 per cent rate of return on capital may be considered appropriate.

Returns to management consist of administrative expenses, mostly those incurred in operating the Gezira Board and mechanized farms and an imputed tenant income. While the former can be readily obtained from the enterprise’s balance sheet, there is no solid basis for estimating the latter. Tenant income derives partly from his labor in the fields, already accounted for under “labor” costs, and partly from the net profits derived from the selling of his crops. These cannot be taken as a return to his managerial function, as they are a residual distorted by the price and taxation systems. Instead, one can proceed on the assumption that the tenant should derive at least as much income as a relatively skilled worker would in an urban area. For a small tenancy of 20 feddans, he should derive about LSd 200 a year for his managerial functions. With the addition of his labor income—from LSd 140 to LSd 250—he would be earning the equivalent of a low-salaried employee in an urban center. Since the evaluation of competitiveness involves considerable adjustments to the raw data and requires the imputation of implicit costs, errors in estimations are bound to emerge. While most of these tend to offset one another, all estimates of competitiveness should be viewed within a good 10 per cent margin of error.

II. Land and Water Charges for the Gezira Scheme

The provision of perennial irrigation over large areas has always been expensive, and the irrigation schemes along the Blue Nile River in Sudan are no exception. Extensive work was required in infrastructure, first in the construction of the Sennar Dam (1925), then in the construction of the Roseires Dam (in the 1960s), the excavation of canals, the construction of water regulators and drains, and the provision of administrative quarters. The historical costs of these works amounted to LSd 125 million for the Gezira scheme but are of little significance because of the secular rise in prices and because of the difficulty of allocating cost from major hydroelectric works, such as dams, which produce a number of joint products (irrigation, flood prevention, and hydroelectricity) in addition to the ecological impact and social benefits, such as the dissemination of agricultural skills.

Nevertheless, it is useful to attain some order of magnitude of the capital costs that would have been required to develop the scheme at current prices. In 1966, the Rist Report estimated the costs of the Gezira scheme on a replacement basis, adjusted for building the Roseires Dam, at LSd 190 million. 28 To update these costs to the end of 1978, the export-price index of industrial countries was chosen for lack of a better index, since much of the equipment needed for developing such a scheme would have been imported from them. 29 Over the period 1966-78, an annual rate of price increase of 9.4 per cent was derived, which brings the cost of the Gezira scheme by the end of 1978 to LSd 512 million, or LSd 320 per cropped feddan. 30 This estimate of the current value of the assets of the scheme (on a replacement basis) is below that for the recently developed Rahad scheme, whose cost of at least LSd 150 million would amount to LSd 500 per feddan but whose potential yields are significantly higher than in Gezira.

Since the Government levies heavy taxes on agricultural crops, the issue of land and water charges cannot be separated from government taxation. Instead, it should be related solely to the value of the government-owned assets and the agricultural surplus obtained. Once the government share of the surplus is determined, it then becomes a matter of policy as to how this share should be apportioned between land and water charges and taxes. The question then arises as to the rate of return that the Government should expect to obtain from these assets. Had the Government operated the Gezira scheme as a state farm with hired labor, its return for the scheme would amount to its true rental value, namely, the residual to capital and management from the output value after deduction of costs of production and depreciation. But the Gezira scheme is cultivated by tenants who are in fact sharecroppers in partnership with the Government and the Gezira Board. While part of their income is derived from their labor, they also get a share of the residual in compensation for their management.

With a full detail of costs of production and output values (Table 4), the derivation of a residual to capital and management is a straightforward matter. However, to each cropping pattern will correspond a different residual. The derivation of an expected profit share accruing to the Government must be carried out in terms of an efficient utilization of its fixed assets, that is, in terms of an “optimum” cropping pattern as derived in Appendix III. Under such a cropping pattern, and with output measured at world prices, the gross output value of the Gezira scheme would have amounted in 1976/77 to about LSd 160 million, while input costs would have been roughly LSd 120 million. This would leave about LSd 40 million to be divided between tenants and the Government. Returns to tenants have been estimated at LSd 19 million (returns to tenant management in Table 4), leaving about LSd 21 million for the Government to collect as land and water charges or under any form of indirect taxation. However, unless indirect taxes are distributed among crops in line with their use of resources, it would be more efficient to apportion the government share among crops as land and water charges, rather than to distort farmers’ incentives through indirect taxation. It is therefore suggested that the entire LSd 21 million be distributed in land and water charges. 31 This distribution of the residual between tenants and the Government is in line with the cotton profit-sharing formulations in effect in the scheme over the past two decades and, hence, with what the Government visualized the tenants’ profit share to be. Looking at it from another angle, this return to the Government would amount to 5 per cent of the estimated asset value in 1976/77. Considering that this rate of return is obtained over and above a built-in 9 per cent rate of appreciation of the scheme’s assets (owing to inflation), it appears to be quite adequate. However, with the 1978 devaluation, the potential surplus available from the Gezira agricultural product would rise by about LSd 9 million, raising the opportunity for some indirect taxation as well as for an improvement in tenants’ incentives through higher income.

The distribution of the Government’s profit share among crops must take into account the watering requirement of each crop and the time during which it occupies the land. As cotton requires the largest number of waterings and occupies the land the longest, its land and water charges for 1976/77 may be set at LSd 18 per feddan. Charges per feddan for other crops would be as follows: wheat LSd 12, groundnuts LSd 10, sorghum LSd 8, and rice and vegetables LSd 14. 32 To these charges must be added depreciation costs, water delivery operating costs, and the administrative costs of the Gezira Board. Present depreciation costs are quite difficult to determine for Gezira, given the age structure of its assets. Depreciation of the agricultural machinery used is included in its rental charges imposed on the tenants, while the depreciation of other durable assets (dwellings, roads, lighting systems, and railways) is included in the administrative costs of the Gezira Board. Similarly, the Ministry of Irrigation does include depreciation costs in its budgeted expenditure. Whether these depreciation provisions have been adequate cannot be ascertained without detailed information on the age profile of the fixed assets and their replacement costs. Consequently, they are taken here at face value and are included in the administrative and irrigation costs shown for 1976/77 in Table 4. In that year, the administrative expenses of the Gezira Board amounted to about LSd 4.7 million. Since cotton assumes a disproportionate share of these costs—requiring much greater supervision and more services than the other crops—it has been assigned an administrative cost of LSd 5 per feddan versus LSd 2 per feddan for the other crops.

After protracted discussions on the desirability of imposing land and water charges on the Gezira scheme, the Government announced in March 1979 that land and water charges are to be imposed on Gezira crops other than cotton and sorghum beginning with the 1979/80 season. The profit-sharing formula on cotton is to be maintained with the addition of an incentive formula whereby the tenants would reap progressively higher shares of profits with higher yields. Previously, the joint profit sharing by the three partners rewarded the inefficient tenants at the expense of the efficient tenants. While the major costs of production (plowing, ridging, fertilizing, pest control, and uprooting) are basically the same across tenancies of equal size, the costs are charged per kantar of cotton produced. Thus, low-yield tenancies were bearing lower shares of costs. Moreover, the profit-sharing formula, akin to a uniform income tax, deprived the tenant from the full benefits of higher productivity and yield. This arrangement was being justified on grounds of income distribution, since it was argued that land fertility varied widely across tenancies. The introduction of a progressive formula attempts to protect the low-yielding farmer by ensuring him a minimum profit share of 49 per cent while giving the efficient farmer additional incentives to raise his yields. However, the new formula is quite complex, and it remains to be seen whether it can be implemented effectively. Sorghum was being exempted, as it is intended to phase out its cultivation in Gezira; however, given the tenants’ past resistance to any reduction in sorghum acreage, its exemption from land and water charges is only expected to strengthen their resolve.

The announced Gezira charges are shown in column 5 of Table 6, together with the 1976/77 charges updated to 1979/80 (column 4) as suggested in this paper. For purposes of comparison, land and water charges estimated for the Rahad Corporation have also been included (column 6). The charges enacted for the Gezira scheme have been derived by the Ministry of Agriculture by estimating water delivery and administrative costs and distributing them among crops. They do not include any return on capital and management, leaving the profits of crops other than cotton almost entirely to the tenants. Prior to July 1979, a large tax load was also imposed on cotton cultivation, with the result that its total charges were much larger than the rate suggested here and than those proposed in the Rahad scheme. This approach tended to burden cotton with a disproportionate share of the agricultural surplus siphoned off by the Government instead of apportioning the Government’s profit share among crops strictly in terms of land and water use. And, while the imposition of land and water charges in Gezira can be seen as a major step in recognizing that cultivators must bear the full costs of cultivating all crops, the Government’s share from the Gezira’s agricultural surplus continues to stem almost exclusively from the cultivation of cotton, thus maintaining a substantial disincentive against its cultivation. While the tax load on cotton has been greatly reduced by applying to it the same exchange rate as to other crops and by abolishing its export tax, the next step would be to reapportion the Government’s surplus share among all crops in terms of their actual use of land and water resources.

Table 6.Sudan: Estimated and Announced Land and Water Charges(In Sudanese pounds per feddan)
CropReturn on Capital and Management (1)Water Delivery Costs 1 (2)Administrative Costs 2 (3)Suggested Land and Water Charges, 1979/80 3 (1 + 2 + 3) (4)Announced Land and Water Charges for 1979/80 (5)Suggested Rates for the Rahad Scheme (1977) 4 (6)
Cotton21.56.56.034.030.3 534.2
Wheat14.55.52.422.56.528.5 6
Groundnuts12.04.42.419.05.522.8
Sorghum9.53.22.415.0--
Rice17.08.72.428.09.0-
Vegetables17.08.72.428.09.0-
Sources: Gezira Board, Ministry of Agriculture, and the Rahad Corporation.

Based on estimates by the Ministry of Irrigation of LSd 0.451 per watering in 1977.

Estimated from the Gezira administrative budget.

Updated from charges for 1976/77 shown in Table 4 by applying the increase in the export price index given in Table 2.

Rates derived in study commissioned by the Rahad Corporation.

Imputed as follows: Government’s profit share, LSd 11.5 per feddan (actual payment by the Gezira Board to the Government in 1977/78 of LSd 6 million divided by 530,000 feddans; Gezira Board’s administrative costs, LSd 5.3 per feddan; LSd 26 per feddan as the implicit export tax on cotton lint; LSd 9 per feddan from a 5 per cent development tax; LSd 4.5 per feddan from the 2 per cent tax for the National Support Fund; and a price support fund tax of ½ of 1 per cent. Taxes on imported production inputs not included. Prior to July 1979, imputed land and water charges were LSd 56.3 per feddan, as cotton was subject to the official exchange rate, equivalent to an implicit export tax of LSd 26 per feddan. The addition of indirect taxes to the charges of the other crops would raise them only marginally.

The Rahad Scheme is not supposed to grow wheat; hence, this rate was imputed on the basis of its water requirements. Actual charges for 1979/80 are LSd 1.50 per watering. Assuming 12 waterings for cotton and 8 waterings for groundnuts, these charges would amount to LSd 18 per feddan and LSd 12 per feddan, respectively.

Sources: Gezira Board, Ministry of Agriculture, and the Rahad Corporation.

Based on estimates by the Ministry of Irrigation of LSd 0.451 per watering in 1977.

Estimated from the Gezira administrative budget.

Updated from charges for 1976/77 shown in Table 4 by applying the increase in the export price index given in Table 2.

Rates derived in study commissioned by the Rahad Corporation.

Imputed as follows: Government’s profit share, LSd 11.5 per feddan (actual payment by the Gezira Board to the Government in 1977/78 of LSd 6 million divided by 530,000 feddans; Gezira Board’s administrative costs, LSd 5.3 per feddan; LSd 26 per feddan as the implicit export tax on cotton lint; LSd 9 per feddan from a 5 per cent development tax; LSd 4.5 per feddan from the 2 per cent tax for the National Support Fund; and a price support fund tax of ½ of 1 per cent. Taxes on imported production inputs not included. Prior to July 1979, imputed land and water charges were LSd 56.3 per feddan, as cotton was subject to the official exchange rate, equivalent to an implicit export tax of LSd 26 per feddan. The addition of indirect taxes to the charges of the other crops would raise them only marginally.

The Rahad Scheme is not supposed to grow wheat; hence, this rate was imputed on the basis of its water requirements. Actual charges for 1979/80 are LSd 1.50 per watering. Assuming 12 waterings for cotton and 8 waterings for groundnuts, these charges would amount to LSd 18 per feddan and LSd 12 per feddan, respectively.

III. Toward an Optimum Cropping Pattern in the Gezira Scheme

The Gezira scheme, comprising 2.1 million feddans (possibly the largest centrally managed agricultural farm in the world), is the center of Sudan’s agricultural activity and accounts for the bulk of its cotton production. It is important not only because of the magnitudes involved but also because it has served historically as a model for the remaining schemes under irrigation. A major yardstick in its performance would be its net foreign exchange benefits, either as revenues earned from its export crops or as savings reaped by the cultivation of import substitutes. Given the costs of production of the various crops, their world prices, and various constraints on production, their foreign exchange benefits as shown in line 6.b of Table 4 can be maximized by the choice of an optimum cropping pattern.

Crop distribution in the Gezira is constrained by the water delivery capacity of the canals. Of the total area of the scheme, no more than 1.2 million feddans can be irrigated at any one time. The seasonality of crop distribution shown in Chart 4 reveals the contraints governing land utilization. From it, one can surmise several facts:

Chart 4.Sudan: Crop Seasonality in the Gezira Scheme

(1) Because cotton occupies the land almost twice as long as other crops, it competes with all of them. Should its acreage attain the limit of 1.2 million feddans, no other crop could be cultivated in the Gezira. Any acreage below this limit that is not used for cotton can be planted to either sorghum or groundnuts in the summer or wheat in the winter. Conversely, at full-capacity utilization, the expansion in cotton cultivation by one acre would entail the reduction in sorghum (or groundnuts) and wheat cultivation by one acre each.

(2) Under present sowing and harvesting practices, all crops overlap in October; during this period, all of them compete for water. This puts great strains on the water delivery capacity and adversely affects the yields of those crops that are denied their water requirements.

(3) The overlapping of crops in October notwithstanding, sorghum and groundnuts compete directly. Any reduction in sorghum acreage can be filled in by groundnut cultivation; wheat competes only with cotton. This seasonality provided a case for wheat cultivation. It was argued that any land planted to sorghum and groundnuts during the summer would have a counterpart during the winter, which would be either left fallow or planted to a winter crop. Cultivation of wheat would make use of fallow land and machinery that would otherwise be idle at that time of year.

With no overlapping among crops and under assumptions of constant returns to scale and of abundant labor, the relative benefits of the various crops for 1976/77 (Table 4) would have prescribed that cotton be cultivated in Gezira up to the limit provided by the availability of water (1.2 million feddans), while the remaining land would be kept fallow. 33 However, such assumptions would not do justice to the problem, since land in Gezira is quite heterogeneous, causing yields to vary significantly with the scale of production and the availability of labor. Hence, cotton cultivation would have to expand up to the point where its marginal revenue product would fall below the combined marginal revenues of wheat and groundnuts. Moreover, the overlapping of crops in October has also caused major variations in yield. The expansion of wheat in 1974/75 and 1975/ 76 from 150,000 feddans to 600,000 feddans (with a concomitant decline in cotton acreage) was accompanied by a sharp drop in the yield of both wheat and cotton. With the subsequent retrenchment of wheat acreage over the following two years and an expansion in cotton acreage, the yield of both cotton and wheat improved substantially. It is not suggested here that distribution of acreage between wheat and cotton is the only factor affecting their yields. Acreage variations of groundnuts and sorghum also have their effect, as well as such exogenous factors as weather (which may affect some crops more than others), the incidence of pests, or the timely provision of inputs. Nevertheless, a striking correlation between yields and acreage variations for these two crops appears to have emerged since these sharp variations in acreage were introduced. As wheat is ecologically ill-suited to Sudan (partly because the winter season is so short), it becomes very sensitive to sowing dates and to the availability of water. Its cultivation beyond 400,000 feddans seems to affect its own yield adversely and that of cotton. On the other hand, it is capital (and therefore import) intensive and does not run into the labor constraint that is manifesting itself increasingly in Sudan.

From the preceding discussion, the major lines of a cropping configuration that would be consistent with the relative crop benefits and the constraints governing their cultivation emerge as follows:

(1) There is no economic justification for sorghum cultivation in Gezira, since groundnuts, which compete with it directly, are a much more profitable crop. Thus, about 300,000 feddans of the 344,000 feddans presently under sorghum cultivation should be released for either cotton or groundnut cultivation. The remaining 44,000 feddans of sorghum would be cultivated on marginal lands not suited to other crops.

(2) It would be more profitable to expand cotton cultivation on the 300,000 feddans released by sorghum than to expand groundnut cultivation, which is more efficient in rainfed lands. But an expansion in cotton would entail a reduction in both a summer crop (sorghum) and a winter crop (wheat). Thus, should cotton acreage increase from 500,000 feddans to 800,000 feddans, wheat would be reduced from its 450,000 feddans to 300,000 feddans, in addition to the reduction of sorghum. Nevertheless, net foreign exchange earnings would increase. The configuration suggested here and the resulting change in net foreign exchange earnings would be as follows:

CropCropping Pattern, 1978/79Suggested Cropping PatternChanges in Total AcreageChanges in Net Foreign Exchange Earnings

Million U.S. dollars
(Summer)(Winter)
Sorghum350,00050,000-300,000-11.9
Groundnuts260,000260,000
Wheat450,000300,000-150,000-4.634
Cotton500,000800,000800,000300,000+71.435
Total cropped area (in feddans)1,560,0001,110,0001,100,000-150,00054.9

A shift from the 1978/79 cropping pattern to the one proposed here would increase net foreign exchange earnings by US$55 million a year. 35 By keeping the cropped area below the theoretical maximum of 1.2 million feddans, the strain on the water delivery capacity in October would be reduced and would provide the authorities with greater freedom for marginal adjustments. 36 The envisaged expansion in cotton acreage would be mostly in the medium-staple variety (in which Sudan is a price taker), which is suitable for mechanical harvesting, given the existing labor constraints.

Mr. Nashashibi, Senior Economist in the Middle East Department, received his doctorate from the University of California at Berkeley. Before joining the Fund, he was with the Center for Development Planning, Projections and Policies at the United Nations. He is the coauthor of Foreign Trade Regimes and Economic Development: Egypt, National Bureau of Economic Research (Columbia University Press, 1975).

See, for instance, Edward Bacha and Lance Taylor, “Foreign Exchange Shadow Prices: A Critical Review of Current Theories,” in Analysis of Development Problems: Studies of the Chilean Economy, ed. by Richard S. Eckhaus and P. N. Rosenstein-Rodan (New York, 1973).

Lawrence H. Officer, “The Purchasing-Power-Parity Theory of Exchange Rates: A Review Article,” Staff Papers, Vol. 23 (March 1976), pp. 1-60.

United Nations Industrial Development Organization, Profiles of Manufacturing Establishments, ID/SER.E/6 (New York, 1971).

The industrial sector is relatively small, accounting for only 10 per cent of the gross domestic product (GDP), and is comprised mostly of agricultural processing activities.

One feddan equals 1.04 acres equals 0.42 hectare.

These issues were discussed in detail in Karim Nashashibi, “The Taxation of Agricultural Crops in the Gezira Scheme and the ‘Land and Water Charge’ Proposal” (mimeographed, 1976).

Realized prices of petroleum products increased by about 800 per cent over this period, but the share of petroleum products in Sudan’s total imports has hardly exceeded 10 per cent even in the most recent years. Consequently, a correction of the import-of-manufactures price index for Sudan would not significantly alter these results.

These are “luxury” consumption items that could be sharply reduced or substituted for by domestic production: all drinks and tobacco, perfume and cosmetics, footwear, sporting goods, refrigerators, air conditioners, ovens, passenger cars, spare parts for bicycles and automobiles, and ready-made textile articles. They amounted to LSd 14.1 million in 1970 and LSd 44.1 million in 1978. Part of this increase in imports was realized by means of the exchange that has been obtained by Sudanese residing abroad (“own exchange”) who have been issued a “nil-value” import license. The implicit rates of exchange for these imports reflected a substantial depreciation of the Sudanese pound.

Hollis Chenefy and Moises Syrquin, Patterns of Development, 1950-1970 (Oxford University Press, 1975), Figure 6, p. 37.

Under the 1972 devaluation, the official exchange rate of LSd 1 = US$2.87 was depreciated to an effective rate of exchange of LSd 1 = US$2.50 by means of a tax subsidy scheme of 15 per cent on all external transactions except proceeds from exports of cotton lint and gum arabic. In 1978, the official exchange rate was set at LSd 1 = US$2.50, while the effective exchange rate was depreciated to LSd 1 = US$2.00 by means of taxes and subsidies for all external transactions except proceeds from exports of cotton lint. On September 16, 1979, the tax subsidy system was abolished and the official rate of exchange was set at the previous effective level, i.e., LSd 1 = US$2.00. Additionally, a parallel rate of exchange was established at LSd 1 = US$1.25 for nontraditional exports (about 8 per cent of the total) and for about 30 per cent of imports that were imported previously mostly under the nil-value license system, as well as some invisibles. Effectively, this amounted to a further depreciation of about 5 per cent.

Michael Bruno, “The Optimal Selection of Export-Promoting and Import-Substituting Projects,” Ch. III in United Nations, Planning the External Sector: Techniques, Problems and Policies, ST/TAO/SER.C/91, Sales No.: 67.II.B.5 (New York, 1967), pp. 88-135. See also Anne 0. Krueger, “The Role of Home Goods and Money in Exchange Rate Adjustments” (University of Minnesota, 1969).

Competitiveness was evaluated for 14 crops and 10 industries for Egypt in an assessment of the impact of the exchange regime on the allocation of resources. See Bent Hansen and Karim Nashashibi, Foreign Trade Regimes and Economic Development: Egypt, National Bureau of Economic Research (Columbia University Press, 1975).

However, the opposite is not necessarily true, since in 1972/73 only 30 per cent of cotton acreage was preceded by groundnut cultivation.

Sorghum stalks also provide fodder for the tenants’ cattle and for those of migrant cotton pickers.

The lower efficiency of the other existing sugar operation at Guneid is attributed to much lower yields of cane and to lack of control by the sugar factory of the cane-growing lands, which are owned and cultivated by the area’s farmers.

At 1976/77 prices, which did not depart significantly from average prices over this period.

Rapid shedding of the sesame seeds at maturity limits the harvest time to about two weeks, placing great demand on timely availability of labor.

The inconsistency between the cropping and incentive patterns would have been entirely possible, since incentives in Sudan have resulted from a series of price and tax policies enacted on an ad hoc basis with different, and often conflicting, objectives in mind.

While for cotton and groundnuts, producer prices do reflect actual export prices obtained, they are distorted by implicit and explicit taxes and subsidies.

Tenants in Gezira are strongly organized in a powerful union that negotiates with the Board and the Government on its real profit share (after taxes). However, the labor that they hire for most of the field work is unorganized.

New productive capacity of 600,000 tons will gradually be attained over the next few years, with about one half destined for export. This, however, is not expected to materialize before 1982. Capital costs per ton of sugar produced at the Kenana plant would be about US$2,000. In contrast, at the estimated replacement costs of the Girba plant, the capital cost per ton of capacity would be US$850.

Even in the northern provinces where traditional wheat farmers obtain yields that are 50 per cent higher than in Gezira, it has become profitable over the past three years to cut back on wheat acreage in favor of growing vegetables and fruit.

Even if one were to estimate supply elasticities on the basis of historical data, these would be derived as a function of entirely different levels of output, type of technology, transportation facilities, and input availabilities from those that are expected to prevail over the next few years. Consequently, such coefficients would have little predictive value.

Food and Agriculture Organization, “Perspective Study of Agricultural Development for the Sudan,” ESP/AGS/PD/SUD/73.

The recent abolition of importation under the “nil-value” license system may contribute appreciably to this shift.

This constraint may be relaxed either by a gain in the terms of trade or by increased factor productivity.

Sudan has one of the lowest sources of government revenues from direct taxes, as it ranks thirty-fourth among 35 developing countries for the period 1973-75. See Alan A. Tait, Wilfrid L. M. Gräz, and Barry J. Eichengreen, “International Comparisons of Taxation for Selected Developing Countries, 1972-76,” Staff Papers, Vol. 26 (March 1979), pp. 123-56.

International Bank for Reconstruction and Development, “Gezira Study Mission,” Annex VI (October 1966); Ministry of Irrigation and Hydroelectric Power.

Of the 2.1 million feddans in the scheme, about 1.6 million are cropped. (See Appendix III.)

Under the actual cropping pattern in 1976/77, the residual would amount to only about LSd 30 million. If the Government were to siphon off LSd 21 million as its profit share, it would leave only LSd 9 million to the tenants. But the corollary of nonoptimum cropping patterns in Gezira is a set of domestic crop prices that ensures that the residual is sufficiently large to keep tenants in production. Thus, subsidization of wheat production is a transfer of income from other government revenue sources to the Gezira tenants. Under practices that preceded the imposition of land and water charges, the residual was divided as follows: Government share: Profit share from cotton revenues (36 per cent) plus indirect taxation of agricultural products (e.g., explicit and implicit export taxes on cotton), development taxes on export crops and imported inputs, and customs tariffs on agricultural imports minus the subsidization of the production of wheat and rice, the writing off of some tenants’ debts, the financing of the annual deficits of the Gezira Board, and the financing of the water delivery costs via the Ministry of Irrigation. Tenants’ share: Profit share from cotton (49 per cent) after government taxation. Full profits from the cultivation of wheat, rice, sorghum, and vegetables (including subsidies) minus the taxation of agricultural inputs and export crops.

Cotton requires 12-16 waterings, wheat 10 waterings, groundnuts 8 waterings, sorghum 6-7 waterings, and vegetables and rice 16 waterings.

As the benefits of groundnuts exceed those of sorghum, the choice is essentially between growing cotton over 1.2 million feddans and growing a combination of groundnuts and wheat over 1.2 million feddans each (or a combination of the two possibilities). However, since the net foreign exchange benefits of the former (US$213 per feddan in Table 4) exceed the sum of the latter (US$192 per feddan), a monoculture of cotton would be prescribed.

Assuming that foreign exchange is paid for wheat imports at going world prices. Imports through concessionary terms (e.g., U.S. Public Law 480) would reduce this amount and raise the opportunity cost of growing wheat.

Assuming an expansion of 100,000 feddans in long-staple cotton and of 200,000 feddans in medium-staple cotton.

In the summer of 1975/76, the total cropped area reached 1.156 million feddans, the closest the Gezira scheme had come to the limit of 1.2 million feddans.

Other Resources Citing This Publication