|CERES No. 111 (FAO Ceres, 1986, 50 p.)|
(Growing consumer demand for the temperate zone cereal poses some tricky choices about the advisability of import substitution)
by Derek Byerlee and Jim Longmire
In our article "Food pricing policy and the wheat import trap" (Ceres, Sept.-Dec. 1985), we highlighted increasing wheat imports by Third World countries. Growth of these imports has been especially rapid in the tropical countries lying entirely or largely between latitudes 23°S and 23°N, where wheat, a temperate crop, has not traditionally been grown except on elevated land. (In this article we consider only these countries; thus Brazil is included, but India is not.) The consumption of agricultural products, such as wheat, imported from the temperate (and largely industrialized) countries is to some extent to be expected with increasing interdependence through world trade and a natural desire to diversify diets. After all, many tropical foods, such as spices and beverages, have long been imported and widely consumed in countries. What is unique about wheat consumption in the tropics is that wheat products, usually bread, have become a basic food staple for many people. Consequently, many countries concerned about increasing dependence on imported wheat have shown interest in the possibilities of producing wheat to substitute for imports. Several of these countries, among them Nigeria, Ecuador, Thailand, and the Philippines already have well-established research and development programmes for local wheat production.
In this article, we offer a general overview of trends in wheat consumption in the tropics and examine the economic issues in establishing a domestic wheat industry. We want to address a number of questions. What factors are responsible for the rapid increase in wheat consumption in the tropics? How can we determine whether wheat production is an economic use of a country's scarce resources? Will domestic wheat production actually save foreign exchange by substituting for wheat imports? What other considerations should be included in the decision to embark on a local wheat industry?
Why consumption increases. The countries between 23°S and 23°N embrace most of sub-Saharan Africa, Southeast Asia and the Pacific, and Latin America from Guatemala to Brazil. They produce less than 2 per cent of the wheat in the Third World (most of it in the more temperate southern part of Brazil and in eastern and southern Africa at higher elevations). In recent years, this group of countries has imported around 20 million tons of wheat, or more than 80 per cent of their requirements (see Table 1). Thirty-eight tropical countries consumed more than 100 000 tons of wheat each; nine of these countries consumed one million tons or more annually.
The growing importance of wheat as a staple food for different regions of the tropics is shown in Table 2. Per caput annual consumption of wheat varies from less than 5 kg in Thailand to more than 100 kg in Cuba. Although consumption is still relatively low in sub-Saharan Africa and tropical Asia, the growth rate of per caput consumption in these regions has averaged 4.5 per cent annually over the last two decades. Much of this increased consumption has occurred in urban areas where per caput consumption now exceeds 40 kg in several countries (e.g., Malaysia, Nigeria, and the Sudan).
A number of factors converge to promote wheat consumption in the tropics.
Incomes and urbanization. The most important factor in increased wheat consumption is rising per caput incomes and urbanization. Wheat consumption increases rapidly with income, but then appears to level off at about 50 kg per caput per year for countries, such as Venezuela, with an annual per caput income near $3 000. Wheat consumption in tropical countries began in urban areas, usually by middle- to high-income groups, but over time has spread to poorer groups and smaller towns. Increased wheat consumption in most cases has come about through substitution of wheat for coarse grains and roots and tubers. These changing consumption patterns reflect the preferences of urban consumers for bread as a convenience food to eliminate time and fuel in food preparation.
The concentration of wheat consumption in urban areas also reflects supply factors. With rapid growth of urban areas (often over 5 per cent a year in African countries), lagging domestic production and inadequate marketing infrastructure for local food staples, and plentiful supplies of wheat in world markets, there has been a natural tendency to import wheat to feed urban consumers. This is especially true in countries where large cities are located on the coast.
Low bread prices. Many countries in the tropics have maintained low prices of bread relative to local food staples. Consumer subsidies for bread and importation of wheat at overvalued exchange rates have led to declining real prices of bread and increasing wheat consumption (see our 1985 article).
Food aid. Most food aid to tropical countries is wheat, and although food aid has declined in importance relative to commercial food imports, it is still a major source of wheat imports in a number of tropical countries (e.g., Sudan and Sri Lanka). In addition, many countries that have historically received substantial amounts of food aid are now large commercial importers of wheat (e.g., the Andean countries). Food aid appears to have promoted wheat consumption through establishment of a local wheat-processing industry, development of consumer tastes for wheat products, and implementation of low consumer prices.
Policy alternatives. Interest in wheat production in the tropics reflects a desire by most countries to promote food self-sufficiency. Many governments have seen rapid increases in foreign exchange expenditures for wheat imports as an opportunity to save foreign exchange and at the same time promote domestic agricultural production. In our view, however, domestic wheat production should be analyzed as only one of a number of policy alternatives for reducing wheat import dependence. These include:
1. Food pricing policies In many tropical countries the most effective means of reducing wheat imports is correction of price policy distortions, especially those that lead to low prices of wheat relative to local staples.
2. Import policy and food aid. Food imports and food aid should give more attention to cereals other than wheat, especially maize, which are more compatible with local production potential.
3. Convenience foods based on local food staples Little effort has been made to develop methods for preparation of coarse grains and roots and tubers to meet the preferences and convenience needs of urban consumers. Composite flours, which mix wheat flour with local food staples, also have much potential as long as price incentives favour the substitution of local staples for wheat. Wheat imports can also be reduced in the short run by using a higher flour extraction rate in the milling of wheat and promoting consumption of wholemeal bread.
4. Increasing domestic agricultural production. Alternatives include promoting export crops to generate foreign exchange for importing food, and increased production of local food staples to replace wheat imports. These alternatives can best be analyzed in a comparative advantage framework that measures the returns to the nation of each of these strategies.
Measuring comparative advantage. A simplified example of calculations of comparative advantage is shown in Table 3. In this example, using a standard measure of profitability of a crop based on prices paid and received by farmers, it is more profitable for farmers to produce wheat (an import-substitution crop) than cotton (an export crop). To measure comparative advantage, prices that reflect true costs and returns to the nation are used to calculate profitability.
The steps in this calculation are as follows:
1. The value of locally produced wheat to the nation is the cost of importing that wheat adjusted by costs of transporting local wheat to the consuming point. Farmers receive a higher price because of a tariff on imported wheat. Likewise, the value of cotton to the nation is computed using the export price of cotton. An export tax on cotton in the example is a benefit to the country although it is not included in the farm price.
2. The value of all inputs in wheat production which are normally treated in world markets (e.g., fertilizer and machinery) is the import price (export price if the country exports the input) plus transport costs from the port to the farm. In the example in Table 3, a subsidy of SO per cent on these inputs must be included in the costs to the nation of wheat and cotton production.
3. The returns to domestic resources (non-traded resources of labour, capital, land, and water) is the value of wheat at world equivalent prices less the value of inputs used in wheat production, again at world prices ($160/ ha in the example). To measure comparative advantage, these returns are compared with returns to the use of these resources in other production activities In the example, because the returns to the nation of allocating land to cotton are higher than for wheat, we can say that cotton has a comparative advantage. Hence while wheat is profitable to farmers, there will be a net loss of $140/ha in national income and in foreign exchange if wheat is grown in place of cotton. The objective of growing wheat domestically to save foreign exchange is therefore defeated.
A number of factors determine the comparative advantage of wheat in any given situation:
- The productivity of wheat. In the above example, wheat is more profitable to the nation than cotton at wheat yields over 2.8 ton/ha (assuming no change in costs). Planners interested in an efficient wheat industry must assess the probability of obtaining these yields.
- The productivity of alternative enterprises In the above example, if wheat is to be grown in place of fallow land (because scarcity of water does not allow other crops, such as cotton) then wheat can be profitable to the nation at yields of only 1.7 ton/ha.
- The technology used in wheat production. Use of capital-intensive technology, such as large-scale irrigation structures, tractors, and combine harvesters reduces wheat's comparative advantage in most tropical countries.. Subsidies such as cheap credit may, however, make wheat profitable to farmers.
- The location of wheat consumption and production in relation to the port In some countries (e.g., Thailand and Nigeria), areas with potential for domestic wheat production are located some 1 000 kilometres from the port, which is also the largest city and main wheat consumer. In this situation, local wheat production could have a comparative advantage in substituting for consumption of imported wheat in or near the producing area, but may be quite uneconomic for supplying the port city.
Obviously there is a great deal of variation in potential wheat-growing environments in the tropics. Nonetheless, it is useful to simplify these into four major environments, as shown in Table 4. In all cases the production of wheat is more favourable where the average temperature in January (in countries north of the equator) is below 20°C than in very hot environments where January temperatures average 25°C.
A number of studies of comparative advantage of wheat in these environments have been undertaken recently. These studies enable some preliminary insights into the prospects for domestic wheat production in the tropics.
Highland Ecuador: Ecuador has traditionally produced wheat in the central highlands In recent years, wheat production has declined from 100 000 ha in 1970 to less than 25 000 ha in 1984. Economic analysis shows that wheat in the highlands does have a comparative advantage.
However, the main competing enterprises, especially dairying, are more profitable to farmers because of import protection to dairy products and allocation of subsidized credit to this sector. These policies provide positive rice incentives for dairying, while importation of wheat at an overvalued exchange rate results in negative protection for wheat production. Hence price policy interventions do not encourage highland farmers to exploit their comparative advantage in wheat production.
Northern Thailand: Preliminary estimates show that wheat grown in the cool season after rice or maize, when land is normally left fallow, has a comparative advantage. Even at low yields of one ton/ha, wheat generates an increase in national income of $44/ha. Unlike the Ecuadorian case, price incentives, especially a substantial tariff on imported wheat, are favourable to domestic wheat production and provide incentives for farmers to produce wheat.
Northern Nigeria: A number of wheat production schemes are under way in Nigeria, where a primary objective has been to substitute for wheat imports, which now exceed 1.5 million tons annually. Reasonable yields are possible when wheat is grown in the dry cool season under irrigation. However, given the high cost of the irrigation infrastructure, combined with the capital-intensive nature of technology employed in wheat production, it is not economic for the country to produce wheat locally. Despite substantial price incentives to farmers (e.g., fertilizer and water subsidies and high wheat prices), wheat production has expanded very slowly. Similarly in Senegal, wheat will have potential only as a means of increasing cropping intensity after the high cost of irrigation infrastructure is written off. Even a small-scale labour-intensive wheat scheme in Mali has not been an economic success.
These studies suggest that the best prospects for wheat in tropical countries (outside the traditional highland areas) are in areas with a dry cool season where land is currently left fallow but where sufficient stored moisture is available to support a low-yielding crop of wheat with low levels of inputs. As genetic advances are made in disease resistance, heat stress, and earlines, the range of environments in which wheat can be economically produced in the tropics will slowly expand. It is unlikely, however, that wheat production requiring expensive irrigation infrastructure will ever be economic.
Marketing and utilization. Ironically, given the widespread consumption of wheat in the tropics, a large problem for potential wheat producers is the marketing and milling of the crop. Most mills have been established at coastal cities to process imported wheat, but it is inefficient to transport locally produced wheat long distances to coastal mills. Furthermore, large scale flour millers, accustomed to a well-developed infrastructure and marketing system for imported wheat, are generally not interested in small volumes of variable quality wheat that characterize the early phases of a domestic wheat industry. Hence a new wheat production programme should plan on a capacity for processing and consuming wheat locally. As the volume of local production increases, consideration should be given to establishing small-scale flour mills in the producing region. If the schemes are successful, wheat might eventually be substituted for imported wheat in coastal cities utilizing established large mills. However, wheat production schemes in the tropics that set ambitious short-run objectives of import substitutions are inevitably too ambitious.
The food security perspective. Comparative advantage is only one criterion for deciding on the feasibility of a domestic wheat industry. Many developing countries see dependence on imported wheat from industrialized countries, subject to variation in world prices and supplies, as a threat to their food security. We think this argument is usually not valid. First, we have noted that there may be more efficient ways to reduce dependence on food imports, such as removal of consumer subsidies on wheat, than establishing an inefficient local wheat industry.
Second, in a tropical environment, wheat yields are very sensitive to small deviations from optimal planting dates or high temperatures early in the season. Hence, domestic wheat production is subject to considerable yield fluctuations in the same way that imported wheat is subject to price fluctuations. Food security should be tackled from a broader perspective of stimulating local agricultural production, developing improved infrastructure to market this production, and, possibly, maintaining food reserves.
Investment in a local wheat industry may sometimes be justified by equity considerations. Where wheat has the potential to increase cropping intensity of food-deficit small farmers, food security fo these poorer groups will be improved. The rapid expansion of wheat in Bangladesh and the potential for wheat in highland Burma and Thailand represent this situation and help to justify wheat research and development efforts for poor farmers in these areas. In other cases, the poor may be adversely affected by diversion of resources away from rainfed farming into large-scale irrigation schemes to produce wheat.