Previous - Next
As structural adjustment programs are implemented, the objectives justifying the development of sorghum and millet production and utilization systems need more careful elaboration. Sorghum and millet are still widely viewed as minor, traditional crops in SADCC food systems. They are recognized for their drought tolerance, but the value of these small grains is more commonly seen in their contribution to semisubsistence food supplies than in the commercial market. Even the high value of sorghum and millet in intrarural markets is generally unrecognized. Yet the small grains are essential components of sustainable agricultural systems in the region's extensive semi-arid areas. The development of these areas requires the development of the sorghum and millet economy.
Comparative advantage
The competitive position of sorghum and millet (pearl millet in particular) principally derives from their relative drought tolerance. Sorghum and millet are more suited to areas with high temperatures and low or unstable rainfall than maize. Yield levels ought to be both higher and more stable under conditions of drought. Finger millet also has a competitive advantage over maize in regions with acidic soils.
However the boundaries of the regions in which sorghum and millet perform favourably relative to maize depend on the technologies employed. Currently, maize out-yields sorghum and millet in many, if not most, years in drought-prone regions and it offers higher yields in many areas with acidic soils. This is a result of historical support for improvements in maize productivity. Maize has been favoured by the development and dissemination of improved varieties, by efforts to promote fertilizer use, and by greater extension support designed to improve crop management. Improved technologies have not been widely available for sorghum and millet.
Investment in technology development for sorghum and millet is beginning to result in major improvements in productivity compared with both traditional methods and with maize. Once readily available, these technologies should extend the competitive domain of the small grains. Sorghum and millet currently account for one-guarter of the coarse grains production area in SADCC. With improved technologies, this could expand to 30-50% The economic contributions from the extensive semi-arid regions of southern Africa could significantly improve.
Food security
The most competitive niche for sorghum and millet is in the rural market. Most semiarid production zones where sorghum and millet are widely grown are consistently net food importers. Even in favourable rainfall years, only a small minority of households produce grain for market. Substantially more grain is purchased than is sold. Most households producing sorghum and millet experience frequent shortfalls in household grain production. Many never produce enough grain to meet their annual consumption requirements.
The immediate priority for improving the productivity of sorghum and millet and the use of these crops must be to improve the food security of rural households. In most semi-arid parts of the SADCC region, major production gains will first serve to improve the level and stability of food consumption among rural households. Productivity gains must also be backed by improvements in grain storage and in increasing grain transfers directly from surplus to deficit households.
Sustainability
The agricultural production systems currently used in most semi-arid areas of the SADCC region are not sustainable. Low incomes and food insecurity limit incentives to preserve the environment. Low productivity has resulted in loss of the most educated and skilled members of the agricultural work-force. Remittances are critical in sustaining these meagre production systems and supplementing household food supplies. Investment strategies are correspondingly geared toward attaining a semisubsistence level of production while children are educated enough to take advantage of employment opportunities off the farm. Farmers are much more likely to invest what little money they have in school fees than in fertilizer. Farming investments favouring future generations are simply not perceived as important.
The future of agriculture in SADCC's semiarid regions remains dependent on significant improvements in the productivity of these zones. In the long run, these areas may best be left for extensive livestock production. In the short and medium term, however, the capacity of the SADCC economies to absorb population migration from semi-arid regions will remain limited. Zimbabwe has the largest and, arguably, the most dynamic industrial sector in SADCC. Yet the formal sector in this country can employ only 5-10% of the children leaving secondary school each year. The returns on scarce resources invested in agriculture must rise to attract greater investment in the semiarid production systems. These returns must remain adequate for at least as long as it takes to absorb the households into the larger industrial economy.
Market development
The costs of grain-market controls have proved unsustainable throughout the SADCC region. The existence of broad areas with continuing food deficits and the rising cost of moving grain over long distances justify emphasis on intrazonal grain marketing. Policy should encourage the establishment or strengthening of a broader private network of grain trade, storage, and processing. Policies favouring extraction, centralized storage, and industrial grain processing must be replaced with policies favouring development of competitive intrarural markets.
Efforts to supply grain to industry on a competitive basis should concentrate on specialized crops for which industry is willing to pay a premium and on the more commercialized production zones along rail lines. In general, sorghum cannot compete with maize in industrial uses for which they are close substitutes. Sorghum and millet produced by a large number of small farmers located in outlying regions necessarily incur high assembly and transport costs. Maize tends to be drawn from regions with higher average rainfall and larger and consistent surpluses that are situated closer to major processing plants.
Economic growth
Efforts to promote sorghum and millet production should not be based simply on desire for equity or concern about the welfare of those producing insufficient food. The development of the small grains food (and feed) system should be viewed as a contribution to national economic growth. Policies and investment strategies should be designed to exploit the competitive advantages of these small grains-a basis for improving the productivity of the extensive semi-arid (and acidic soil) regions of the SADCC countries and of their rural labour force. Gains to the economy will also accrue from improving rural food security, reducing the need for drought relief, lowering the level of subsidies underlying grain markets, and, at least in the short run, stemming migration from rural to urban areas.
Strategies for development of the sorghum and millet subsector can be mapped, in part, from an understanding of the current position of these grains in the national food system. In comparison with the strength of the maize economy, the food system for sorghum and millet is primitive. Although supplies of cheap maize meal produced by urban industry are ubiquitous, little sorghum or millet enters the national market. The vast majority of sorghum and millet is consumed by farm households.
The structure of this system is illustrated by an analysis of sorghum grain flow in Zimbabwe, one of the more commercialized of the region's sorghum economies (Figure 1). Small farmers account for over 90% of sorghum production by area and 80% by production level. Only 7% of national sorghum crop is sold through formal sector markets (to the grain marketing board). Of the marketing board's portion, only 75% comes from large-scale commercial farms. Although small farmers produce the most grain, they deliver only a limited amount to the national market. Of the sorghum produced by small farmers, 98% is retained for household consumption and neighbourhood sales.
Figure 1. Production and use of sorghum in Zimbabwe, 1990-91. (GMB, grain marketing board.)
Figure 2. Grain movement in the food system of small farmers in semi-arid regions.
In 1990, sorghum markets were partially liberalized and red sorghum markets were deregulated. As a result, large-scale farmers now sell most of their crop directly to the opaque brewing industry. The principal brewer obtains his entire supply of sorghum for malt domestically. During the 1980s, this single source of demand accounted for over 90% of the sorghum used by Zimbabwe's industries.
During the 1990-91 marketing year (Figure 1), unusually large stocks of sorghum were held by the marketing board as a result of uncompetitive pricing policies. Following the 1982 84 drought, the large-scale commercial farm sector doubled the area of its sorghum plantings. Favourable rainfall during the following 2 years tripled production and market deliveries. The government refused to adjust the sorghum selling price downward and stocks simply accumulated. Five years later, the grain marketing board was finally disposing of the last of these stocks at large discounts (and trade account losses). The old stocks were purchased by the brewing and feed industries. Large discounts were also offered in export markets.
At market determined prices, the brewing industry purchases roughly 17 000 Mt of good malt-quality sorghum each year. This demand is forecast to remain steady. Small quantities of low-quality grain may be sold to the stockfeed industry, but most will continue to be retained for rural consumption.
This picture of the formal market can be extended with a closer look at the food system of small farmers in semi-arid regions of Zimbabwe (Figure 2). Roughly 90% of grain (sorghum, millet, and maize) production in these drought prone regions is retained for household consumption. Of the remainder, 23% may reach national markets through parastatal marketing authorities. The remaining 6-7% is sold through neighbourhood markets-primarily to neighbouring households. Market regulations (and subsidies) in most SADCC countries have prevented the development of private grain trading networks. As a result, virtually no grain flows over long distances through informal channels. Village grain markets are primitive or nonexistent.
Because most semi-arid regions face persistent grain deficits, imports are always needed. Rural households must purchase an average of 20% of their grain each year. This may rise as high as 60% during years of drought.
Virtually all of these purchases are of maize-mainly in the form of industrially produced maize meal. Even when sorghum stocks are large, direct sales back into regions of grain deficit have been limited.
Most marketing boards are simply not oriented toward redistributing grain to rural areas. Further, the private trading network necessary to extend grain distribution beyond marketing board depots is nonexistent. In many villages, industrially milled maize is the only grain available for purchase.
In Zimbabwe, the 20% annual deficit facing semi-arid regions implies a need for almost 100 000 Mt per year of imported grain. This deficit may be offset only by a 40% increase in total sorghum and millet production.
Zimbabwe maintains enough maize in stock to avoid grain imports. However, following the drought in 1990-91, grain deficits ranged from 50-80% of consumption requirements in the driest regions (Figure 3). This implies a need for up to 400 000 Mt of grain-roughly twice the average sorghum and millet production by small farmers.
These relations are similar across all the countries of the SADCC region. In countries with better developed wholesale and retail trading networks, much of the deficit may be offset by industrially milled maize. In regions where grain and meal markets are limited, consumption levels simply decline.
Sorghum and millet remain critically important food security crops for many of the poorest and most drought-prone areas of the SADCC countries. As markets are liberalized and maize meal subsidies are withdrawn the relative position of the alternative coarse grains will shift. As new varieties and hybrids become available, sorghum and millet should begin to play a more important role in the SADCC food system.
The priorities for developing the small grains food system must reflect both technological opportunities and utilization preferences. These forces of supply and demand must be viewed from a dynamic perspective that takes into account changes in the broader macroeconomic environment and in opportunities to exploit an evolving set of production and consumption niches. This evolution can be shaped through:
The major functional components of the food system represent major decision points for investment or resource allocation. They include grain production, storage, processing, trade, and consumption. At many points on the continuum from grain production to consumption, decision-makers must judge the returns to competing investments. They must decide whether to allocate land to sorghum or maize, whether to store these grains for later sale or consumption, whether to process or trade these grains in the formal market. The environment of policy, infrastructure, and technology determines the structure of incentives guiding these decisions.
Investment in the sorghum and millet subsector can be guided by government support and broad agreement regarding development priorities. To date, few countries in the SADCC region have seriously considered what these priorities ought to be. None have articulated strategies for developing semi-arid agricultural systems.
The production system
The dominance of maize in many semiarid cropping systems and in areas more suited to the production of sorghum and millet derives from strong national policies favouring the production of this key staple. Strategies designed to promote expanded grain production in zones of higher potential have simply been extended to the drier and acid-soil regions. Subsector strategies to support the production and use of sorghum and millet must first be distinguished from those underlying the larger cereal grain economy. Specialized development plans are particularly needed for each country's drought-prone regions.
Most households in the SADCC region growing sorghum or millet also grow some maize. Because of the availability of improved technologies, maize is commonly the more productive crop. Sorghum and millet are grown to offset the rinks of severe drought and for specialized use as beer malt, but due to the strength of historical public support and relative productivity, maize is viewed as the commercially important enterprise.
If sorghum and millet are to compete with maize in smallholder production systems and in rural and industrial consumption systems, the productivity of these crops must significantly increase. Productivity gains must offer returns higher than those from maize and higher than the earnings of labour and capital resources off the farm. The value attached to increased productivity will not necessarily be simply grain yield. Farmers most concerned about food deficits may place a higher value on less variability in yields, greater minimum yields, and better storage properties. Returns to labour are also important and farmers have consistently shown their willingness to sacrifice yield for taste. Differing values may also be placed on the relative ease with which alterative varieties can be processed.
Initial productivity gains must be achievable with limited capital investment. Capital is an extremely scarce resource. Farmers in semiarid production systems display a common aversion to risky investment in fertilizer. If credit is available, purchases of livestock may be preferred to purchases of chemical inputs.
The greatest immediate potential for improving the competitive position of sorghum and millet is the development of improved varieties and hybrids. Small farmers in semiarid regions have shown a willingness to try new seeds. Hybrid maize has been adopted by more than 90% of Zimbabwe's small farmers, including many farming in highly drought-prone conditions. Sorghum and millet have a genetic potential to perform significantly better than maize in these regions. Recent experimental station and on-farm trials are providing accumulating evidence of these gains. Further testing is now needed to judge whether the prospective yield gains are associated with the range of additional grain traits that farmers value.
A second major determinant of production decisions about alternative grains is the structure of extension support. Maize promotion programs have led to circumstances where extension workers based in semi-arid regions are much more likely to know the latest recommendations for maize than for sorghum or millet. Extension workers are also unlikely to understand the complexities of risk management strategies, e.g., planting a crop three or four times over several months. They are not taught how to respond to such behaviour. When farmers fail to attend meetings and ignore advice, the incentive to provide additional advice declines. Many extension workers are more likely to accuse farmers of not wanting to learn about new technologies than to adapt their recommendations to the constraints guiding farm decisions.
Finally, as new technologies become available, the physical distribution system is a constraint. Seed companies are less likely to invest in promotion of new varieties when the level and consistency of demand is unknown. Seed distribution is more difficult and costly in outlying regions. When seed prices are controlled, profit motives encourage the concentration of resources in areas along the rail lines. The introduction of open pollinated varieties implies a further risk to the companies. Although an initial market may be significant, future demand may be limited.
Grain versus livestock feed
Many semi-arid areas in the SADCC region receive such low levels of rainfall that they are suited only for livestock. In general, these livestock production systems are also classed as subsistence. Cattle are not raised for commercial markets. They are maintained for draft power, milk, and income security. The value of the services provided by these animals and the value gained through reproduction (although low) tends to be higher than the value of crop production. It is also often higher than the value of the animals if they are sold on the commercial market.
Relatively little effort has been directed toward improving cattle raising methods in small-scale production areas. Farmers are encouraged to use crop residues, but crop breeding programs generally do not measure the relative value of these residues in terns of digestibility, protein content, etc. The importance of residues may be recognized, but the trade-offs between grain yield and residue value are unknown. The need to provide varieties with varying residue values to alternative farm populations is seldom considered.
Most small farmers living in semi-arid regions will continue to grow food crops regardless of their limited productivity. But consideration of the economic future of semiarid production systems must take into account the importance of crop-livestock interactions. Efforts to maximize productivity should assess returns to investments in combined crop and livestock enterprises. Ultimately, these areas may best be viewed as subsistence production systems for crops and as commercial production systems for livestock.
Grain processing
The International Development Research Centre (IDRC) has promoted the introduction of small grain dehullers into the market systems of southern Africa. This strategy is based on the premise that a major disincentive to the production of sorghum and millet is the constraint to the use of these crops resulting from the laborious task of processing. Although some farmers process sorghum and millet through nearby hammer-mills, many persist in hand processing due to the poor quality of hammermilled meal. The difficulty of hand processing has led farmers to plant and consume more maize.
Farmers prefer hand processed sorghum or millet, but labour constraints (or the high value of labour in alternative occupations) discourage the practice. Preference for hammermilled maize over hammemlilled small grains promotes the use of less-drought-tolerant crops. Once dehullers are introduced, mechanical processing of small grains should yield acceptable meal.
The validity of these assumptions needs further investigation. Although farmers appear interested in the use of dehullers, it remains unclear how much they are willing to pay for these services. In particular, will payments fully offset unsubsidized capitaland operational costs of the dehullers?
It is also unclear whether the availability of dehulling services will lead to a significant increase in the consumption of sorghum and millet. These services may easily bring a significant increase in the consumption of dehulled maize. Although evidence indicates a clear demand for the services of dehullers, their impact on production and consumption decisions must be evaluated.
Grain storage
National policies for grain storage tend to concentrate on efforts to maintain a central, publicly administered grain stock. Maintaining these stocks is sometimes used to justify trade controls over a principal grain staple such as maize, but in most southern African countries, the largest grain stock is held by farm households. Most farm households store sufficient grain to meet their consumption needs until the next harvest. Small-farm households also usually keep small strategic stocks to be used if the next year's harvest is poor, particularly in regions with a high probability of drought. Most of these stocks are disposed of during the preharvest period if the crop season appears favourable, but small amounts of grains that can be readily stored may be held for 2-5 years.
Finger and bulrush millet can be stored for extended periods-sometimes up to 10 years. In some areas, there is a direct link between this storability and farmers' production strategy. A minimum stock is maintained over extended periods to offset the risks of severe drought. The area of land allocated to these crops depends on the quantity of grain left in stock. When multi-year stocks run low, more land is planted. When millet stocks are large, more land may be planted to the more risky crop, maize.
Knowledge of the importance of rural grain stockholding strategies is limited. Small programs in some countries are directed toward improving methods of household grain storage. Yet in no case has a national storage policy been established that considers household storage as a key component of the national grain stock. Estimates of grain storage losses in the rural areas are highly variable; although they are rarely less than 15% per year, they are sometimes as high as 50%. This represents a tremendous loss in national food supplies.
Revised national policies can be established that take into account the importance of farm household grain stocks for the national grain reserve. Pricing policies must then reflect the value of maintaining these stocks within the rural areas. Such policies can be viewed as a means of reducing the cost of holding the national stock and as a means of reducing the cost of distributing grain back to farms when rural supplies fall short. Part of the effort involved in maintaining central grain stocks should be redirected toward gathering better information about rural stocks, monitoring them, and encouraging more efficient practices.
Technological advances should be sought that extend the storage life of rural grain supplies. These include improving the quality of storage bins and improving storability of improved varieties. The latter may be a function of grain hardness as well as grain shape or size. In addition, further work is needed on the efficient use of chemicals to preserve grain, including traditional pesticides such as wood ash, eucalyptus bark, and certain grasses. Extension agents should be as prepared to provide training in the use of traditional storage practices as in the use of agrochemicals.
Particular emphasis should be placed on rural grain storage policies and practices relating to sorghum and millet. These grains are grown in regions where the value of improved storage may be highest, i.e., where households are most subject to food shortfalls. Sorghum and millet are commonly viewed as food security crops. Stocks are maintained, particularly millets, as insurance against shortfalls. If the storage technologies and incentives relating to these crops can be improved, the payoff in terms of enhanced food security may be high-higher perhaps than the return to additional investment in a centralized strategic grain stock.
Grain trade
Virtually every SADCC country has initiated some form of grain market liberalization over the past few years. Sorghum and millet trade has been deregulated and commitments by public marketing authorities to act as buyers of last resort have ended. In some cases prices for these grains are still announced, but they are no longer enforced. In others, there are no longer official grain prices.
These adjustments have been stimulated primarily by the determination to reduce budget deficits of marketing boards and to reduce national subsidies for grain trade. Such changes have generally not been accompanied by clear strategies for promoting the development of a competitive private-sector marketing system to take the place of the public support.
New strategies are needed to facilitate the movement of grain from surplus to deficit households through private channels. National grain-trading strategies should be designed to encourage the development of competitive grain trade, not simply along the rail lines, where private investments may have the highest payoff in the short term, but also in outlying areas most prone to exploitative pricing practices.
Major initiatives are being launched in several SADCC countries to improve rural roads. In addition, however, major initiatives are needed to improve transport. Extensive investment is needed in building a rural truck fleet, based in and oriented toward serving outlying areas. Special loans are needed for rural entrepreneurs willing to invest in such facilities. The elimination (or relaxation) of pan-territorial prices will also encourage these investments.
Improved investment incentives are also needed to induce traders to invest in commercial grain (and other goods) storage systems. Pan-seasonal prices eliminate the return on investment in grain storage. Again, such controls must be lifted and maize meal subsidies must be reduced or removed. Commercial loans could be targeted toward the construction of rural grain storage facilities. Small-scale traders should be advised on how such investments can be made most efficiently (e.g., the size and strength of commercial grain storage facilities).
Grain traders must also have access to trading capital. Currently, small-scale private grain traders in Zimbabwe (approved buyers acting as representatives of the grain marketing board) are often criticized for buying grain using lines of credit meant for purchasing merchandise for their retail shops. This practice, however, is a resection of the capital constraints these traders face. Few small-scale rural traders have the capacity to tie up the ZWD 5400 required to purchase only 20 Mt of grain. Many even have difficulty obtaining enough money to purchase and hold 1 Mt of grain. There are no financial arrangements to support such investments.
Restrictions on grain movements and controls on grain prices have severely limited the number of traders with experience in buying and selling large quantities of grain. Trading networks must be built up. Removing market controls will stimulate only limited investment and this will tend to be concentrated along rail lines. Regulatory changes must be accompanied by the provision of access to capital, foreign exchange (for vehicles and spare parts), and advice on small-scale investment opportunities.
Industrial demand for sorghum and millet
Less than 5% of the sorghum and millet produced in the SADCC region is used by industry.2 Virtually all of this is used in the opaque beer brewing industry as a flavouring ingredient or source of malt. Cheaper maize remains the principal input. Sorghum and millet are uncompetitive with maize in the range of other industrial uses for which these grains are close substitutes, e.g., stockfeed.
The small grains will remain uncompetitive as long as their productivity remains lower than that of maize and the assembly and transport costs associated with the collection and movement of grain to the milling or brewing plant are high. The brewing industries in Zimbabwe and Tanzania derive their sorghum largely from large-scale commercial farms close to their breweries. This practice allows a greater degree of control over the quality of the grain as well as low costs.
The likelihood of a significant increase in the use of sorghum (or millet) by the opaque beer industries in the region is limited. Large increases in productivity relative to maize would be needed to justify switching to these grains. If these productivity gains are achieved, this grain might be better used to resolve shortfalls in rural grain supply. The potential for expanding the use of sorghum (or millet) in beer making depends on the strength of the brewing industry in each country. Although there is a potential to increase the demand for opaque beer in some countries, in Zimbabwe demand seems more likely to decrease. As incomes rise, there is a rapid transition to clear beer consumption. Over the past 5 years, consumtion of opaque beer has remained stable, while that of clear beer has risen at an average annual rate of 20%.
Although there is some interest in the use of sorghum or millet in clear beer production, brewers remain concerned about consumer preference for barley-based beers. Inexperience with the use of sorghum and a ready supply of barley further limit interest in the small grains. As in the case of opaque beer, maize is also a cheaper source of adjunct.
Work in Nigeria has shown the technical feasibility of using sorghum in the production of clear beer. However, this does not mean priority should be placed on this opportunity. There may be stronger justification for promoting expanded production of sorghum for alternative grain uses-such as food security of rural households.
A larger and politically important source of potential demand for sorghum is in the production of composite bread flours. Bread consumption is increasing rapidly throughout the SADCC region. None of the countries in the area is self-sufficient in wheat production. All except Zimbabwe import more than half of their wheat supplies; the SADCC countries imported an average of 550 000 Mt of wheat per year in 1986-88 at an annual cost of over USD 85 million. This compares with an average annual import bill for barley and malt of less than USD 9 million.
The substitution of high-quality white sorghum for wheat has been proven feasible and the economics of this substitution seem to be favourable. The principal constraint to the pursuit of this option is the lack of a consistently available amount of consistently high-quality grain. Although there will always be scope for improving the quality of grain available for composite milling, sorghum varieties are already available that can be milled economically. However, the grain is not available in the quantities needed by industry. Where larger quantities of white sorghum have been available (only in Botswana), its quality has been poor or highly variable.
The milling-baking industry must contract with larger commercial producers for particular varieties of white sorghum with clearly specified characteristics. Grain traders should also be alerted about the market for white sorghum of good milling quality.
Little is known about the potential demand for industrially milled sorghum and millet flour. Millingindustries throughout the SADCC region are geared to the use of maize and wheat. Although several millers in Zimbabwe have experimented with the use of white sorghum, only one miller in Botswana has attempted to market a sorghum flour. This has proved difficult due to uncompetitive grain prices (imported maize is cheaper) and the poor and variable quality of the sorghum. Several millers in Zimbabwe state that they do not believe a demand exists, but limited market trials by Environment Development Activities (ENDA) indicate the possibility of an urban demand for sorghum or millet flour.
The largest potential demand for sorghum and millet is from the stockfeed industry. This is one of the most rapidly growing industries in the SADCC region. If sorghum and millet were substituted for most of the maize currently used for feed, the aggregate demand could be five to ten times higher than the demand for sorghum and millet in the brewing and composite milling industries.
The stockfeed industry provides a source of demand for all grain that does not meet the premium quality standards of the milling and brewing industries. To meet this demand, however, sorghum and millet must be priced competitively with maize. This requires a price discount of 5 to 25% depending on the type of stockfeed being produced and the experience of the industry. This again implies significant improvements in sorghum and millet productivity.
