PART 1
DOMESTIC MARKET OPPORTUNITIES

The success and growth in demand for cassava chips and pellets, and cassava starch in non-producing countries suggests that demand opportunities for cassava could be equal, or greater, in cassava producing countries. Much of this demand opportunity exists because there are new and growing industries that could use cassava and cassava products, and/or there are import substitution opportunities. The following chapters in Part 1 of this presentation contain an examination of the opportunities for cassava, cassava flour, cassava starch and cassava chips in cassava producing countries.

The description and analysis of the global opportunities for cassava is based primarily on data available from FAOSTAT website (FAO 99). This global analysis includes data from 64 countries (31 African, 11 Asian, 19 Latin American and Caribbean countries, and three countries from Oceania) that produced more than 10 000 metric tonnes of cassava in 1995^[1]. These 64 countries accounted for 99 percent of world production in 1995.

Continental trends in production presented in Table 4 show that only Africa has had a continued expansion in production over the 34 year period. Asia has had the most dramatic increase in production but its level of production has levelled off since 1990. For Latin America and Oceania there has been a slight decline in production since 1970.

It can be suggested that the differences in continental cassava production trends reflect the different roles of cassava in each continent. The growth in African cassava production is associated with the continent's population growth and the scale of importance of cassava in its 31 studied countries. Latin American and Oceania declines are explained by the growth in consumption of alternative stables. In contrast, Asian growth is primarily in response to the European demand for cassava as energy rich feed ingredients in animal feeds. On the other hand, the levelling of European demand in the nineties is reflected in the plateau of Asian cassava production since 1990.

This initial data suggests that the domestic market has been and continues to offer potential for growth in Africa. This is in contrast to Asia and Latin America where reliance on the domestic market is not likely to provide a growth market for cassava. However, one avenue of growth would be to promote cassava as an import substitution crop. Prime targets for imports substitution could be wheat and wheat flour; maize; maize starch; dextrin, modified starches, glucose, and dextrose.

The following four chapters provide a brief evaluation of the import substitution opportunities for cassava. The chapters also contain case studies that look at specific market development opportunities and how they might be realized.

1
CASSAVA FOODS

Global Perspectives

Historically cassava has been first and foremost a basic staple in cassava producing countries. Although cassava can be prepared in a large variety of ways for human consumption, most localities rely on a limited number of preparation and processing methods. It is a point of interest why processing and preparation methods have not been more widely disseminated and adopted.

The importance, or non importance, of cassava as a human food is portrayed in Table 5. The most striking feature of Table 1.1 is Africa's substantially higher and steady level of cassava consumption (note: these data refer only to those 64 countries of this study that produce more than 10 000 metric tonnes of cassava annually). To express the African consumption level in different terms, daily cassava consumption is 0.4 kilograms per person. This is greater than the daily consumption level of most staples in any part of the world. The message is that cassava is a very important food crop in the 31 African countries being considered.

The Latin American consumption levels, while much lower than in Africa, equally suggests that cassava is an important human food crop. The consumption of 33.14 kg/year of cassava in Latin America is higher than the region's average potato consumption level. However, it is lower than the consumption of wheat and wheat products or maize and maize products (FAO 1997).

Impact of urban population growth to cassava consumption

It is well documented that developing countries have generally had higher population growth rates than developed countries. What is often overlooked is that urban growth rates in developing countries exceed general population growth rates in all the four regions in consideration (Table 6). In fact the 64 studied countries include some of the fastest urban population growing countries in the world.

Because of higher population growth rates and higher levels of cassava consumption the greatest growth potential for the use of cassava as a human food appears to exist in Africa. Equally Asia has shown some signs of growth, while Latin America a slight decline (Figure 1). North America and Oceania are not included in this graph because their growth levels are insignificant when compared to the other three regions.

Applying simple time trends to these data suggests that by the year 2005 African consumption could be approximately 82 million tonnes, Asian 24 million tonnes and Latin American 13 million tonnes. This estimate represents an increase of approximately 21 million tonnes over the 1995 consumption levels, and is close to the Food and Agriculture Organization (FAO) estimate of a 26 million tonnes increase between the early nineties to 2005 (FAO 1997). The FAO study cited was more detailed, including income and demographic factors and covering all cassava producing countries.

Demographic changes provide both a challenge and an opportunity for increased demand for cassava as a human food. On the one hand, individuals migrating to urban areas may have less desire to eat cassava than they did when they lived in the rural areas. On the other hand, the increasing urban population provides an opportunity to sell more cassava if the product is of good quality and competitive price. Domestically therefore, growing urbanization offers opportunities to develop new, or unexploited markets for cassava. The keys to realising the aforementioned market opportunity are:

• continued consumption of cassava by urban dwellers;

• a distribution system that links producers to urban consumers;

• and a consistent supply of good quality products at a competitive price.

These keys may further require the availability of improved infrastructure, new handling and storage technologies, increased production and improved efficiency.

While the growth potential resulting from urbanization is an unknown, several assumptions can be used to provide an estimate of the potential in the year 2005. For this report it is assumed that those people moving to urban areas will maintain their cassava consumption level at the current national consumption levels, and that growth in urbanization will continue at the rate that has existed for the past 10 years. Thus the increase in demand for cassava in Africa in 2005 could be 13 million tonnes or a 16 percent increase over 1995 production, given urban cassava consumption of 146 kg/year and urban expansion at 5 percent per year (Table 7).

This approach was applied to Asia and Latin America, to provide the following estimates of potential market increase owing to urban growth: Asia, 2.8 million tonnes (six percent of 1995 production), and Latin America, 2.4 million tonnes (eight percent of 1995 production).

Clearly the market increase potential in Africa is more substantial. This same analysis^[2] applied at country level data produces Table 8. For nine of the studied African countries, the growth potential by 2005, owing to urbanization, could be an increase of more than 20 percent over 1995 production levels. Only four other countries, Malaysia, Philippines, Dominican Republic and Venezuela, would appear to have such a large growth potential. Nevertheless, the urban market does appear to offer growth opportunity of at least 10 percent for 39 of the countries studied.

Cassava as a human food in Brazil

The previous analysis suggests that the growth potential for cassava owing to urbanization in Brazil might only be 6.9 percent, or 1.76 million tonnes (Table 8). While the percentage is not great, the volume is substantial. The Brazilian case studies, which follow, provide an indication of how the urban markets can provide new and growing opportunities for cassava. The case studies also indicate the importance of developing new products, methods of processing and distributions to market growth.

Those Brazilian markets represent relatively high value added cassava markets with comprehensive distribution systems, and also a high level of product recognition, and are among some of the most progressive food markets in the world.

1.1. Cassava as fast food

"Pão de queijo" or cheese bread

The main cassava-based fast food in Brazil is "Pão de queijo" (cheese bread). This is a type of bread made from sour cassava starch, cheese and eggs. Sour starch is a starch that has been fermented and sun dried, with fermentation lasting anywhere from two weeks to more than a year. The combined action of fermentation and sun drying induces a physical/chemical modification, giving rise to the expanded starch proprieties that resemble extruded starches.

It is possible to find "paes de queijo" (cheese breads) in many Latin American countries, but its consumption is greatest in Brazil, mainly in the Southern and Centre/western regions, and in Colombia, where it is called "pandebono". During the last five years, Brazilian consumption of "pão de queijo" has increased greatly, progressing from a regional product to a nation wide fast food.

Furthermore, the production of "pão de queijo" has changed from its original recipes. In some cases native and modified cassava starches have substituted sour cassava starch. It is also interesting to note that in Minas Gerais State almost every family has its own recipe for "pão de queijo", made with sour starch. Sour starch gives the product a strong acid taste.

The Brazilian sour cassava starch industry has also undergone a change. For example, there has been a decline in production by traditional small scale starch manufacturers in the traditional production states of Minas Gerais and Santa Catarina, while larger more modern companies are expanding in Paraná State.

Consumption and commercialization

In recent years, the consumption of "pão de queijo" has expanded from Minas Gerais to the city of São Paulo and further to the rest of the country. The expansion of consumption could be explained by the product's endorsement by the former Brazilian president, Itamar Franco, who was from Minas Gerais. He required that "pão de queijo" be present at all official meetings. Consequently his administration was called the "pão de queijo" government (Pão de queiuo agora globalizado 1997b).

"Pão de queijo" consumption is seasonal, and is associated with coffee consumption, which is greater in the winter months. The fact that " pão de queijo" is consumed with coffee explains why its consumers are typically adults. Nevertheless, all classes of people, rich and poor, young and old, consume "pão de queijo".

In 1995, Vilpoux undertook a market study of the commercialisation of "pão de queijo" in the city of São Paulo, Brazil. Altogether 74 supermarkets, 60 bakeries and 36 coffee shops were visited. During the study he found that "pão de queijo" was sold in 95% of São Paulo city bakeries and was being sold alone or filled with jam or sausages. In early 1995, the average price of a simple "pão de queijo" was US$0.70 and US$0.95 for a filled one, with a simple product weighing around 100 g per unit. The study also found that on a weekly basis, the average São Paulo bakery was selling about 23 kg of "pão de queijo". All the interviewed bakeries were buying their products from wholesalers (Vilpoux et al. 1995).

By 1998, this typical Brazilian product was being exported to other Latin American countries, such as Argentina and Peru. Brazilian products are also being exported to Japan, the United States and Europe, where they seem to be appreciated (A Hungna e apenas o comeco 1998a).

In Brazil, supermarkets sell a variety of sweet and sour starch mixes (dry pre-mixes, frozen and refrigerated) facilitating the making of "paes de queijo" at home. During the 1995 market survey, dried pre-mixes were found present in all supermarkets, while the refrigerated and frozen products were present in only 20% and 45% of shops and supermarkets respectively. According to the supermarkets' managers, 55% of consumers preferred dry products, 23% frozen products, while only 3% preferred refrigerated products. However, frozen food in Brazil registered the highest consumption increase in 1997 (18.2%), and is gaining space in all supermarkets in recent years, in contrast to the small space it was reserved in 1995 (Novidades na industria de alimentos 1998b).

Similarly, 65% of the supermarkets' managers interviewed in 1995 confirmed that the "pão de queijo" market was growing, an evolution that is probably continuing into 1998 as has been the case with frozen products.

The expanding network of coffee shops

In Brazil, the expanding consumption of "pão de queijo" has had consequences on "pão de queijo" processing and the entire cassava starch industry. "Pão de queijo" is now more than a traditional regional product. It has become a fast food, consumed by all classes of the population in fast-food restaurants and coffee shops. Large groups of coffee shops now exist in Brazil, such as "Casa de pão de queijo" and "Express pão de queijo", which sell fast-food products with coffee, and are specialized in "pão de queijo". The biggest group, the "Casa de pão de queijo" owns 141 shops in the biggest towns of the country, all of them under franchise, and is opening new shops in Argentina.

Alongside the development of consumption in coffee shops, there has been development of frozen products, which facilitate the handling of "pão de queijo" in coffee shops and supermarkets. The two biggest Brazilian companies, the "Forno de Minas" and the "São Geraldo", are growing very fast and are producing monthly shipments of between 300 and 750 tonnes of frozen "pies de queijo" (Pão de queiuo agora globalizado 1997b). "Forno de Minas" started in 1990 with a production of 40 kg per day, reaching 80 tonnes in 1997 (ibid.). Its sales totalled 40 million dollars in 1997, 10 times more than in 1994 and twice as much as in 1996 (A receita da vovo 1997c). The company had since started exporting to Italy and Peru, with the expectation of regular sales in 1998. Exports to Argentina were expected to commence in the second half of 1998 (ibid.). In Minas Gerais State, where "pão de queijo" had its origin, statistics indicate there are 85 producing companies in the city of Belo Horizonte, capital of the State, and 120 more in other parts of the State, most of them informal (Pão de quetuo agora globalizado 1997b).

Coffee shops need products that are easy to cook and handle, without the necessity of adding ingredients. The three main products used by coffee shops, restaurants and bakeries are dry pre-mixes, refrigerated and frozen products. The dry pre-mix is not easy to use because of the necessity to add water, a complication for the shops; instead the product is used mainly in bakeries, where mixing equipment exists. On the other hand, the frozen product is easy to use, but difficult to transport. It is used in big quantities by some specialized coffee shops networks, with the aim to standardize their products. Most coffee shops use refrigerated products in packets of 5 kg.

Constraints in "pão de queijo" production

The entrance of big "pão de queijo" companies have changed much of the industry. It is true that a national company needs a standard product with consistent quality. But these criterions however are very difficult to maintain in "pão de queijo" production because of the low and unstable quality of cheese, and the low and unstable quality of sour starch.

Yet, cheese and sour starch are important ingredients in the "pão de queijo" industry. The cheese industry in Brazil is becoming more concentrated, with improvement of production, and the technology to make good quality cheeses, all year round, is already available in some companies. Sour starch problems are more complex: The fermentation process is difficult to control with consequences on the uniformity of product. Product quality depends on micro-organisms, themselves depending on temperatures during fermentation. Sun treatment is subject to a lot of contamination, making it difficult to use sour starch in modern food companies. Nevertheless the product (sour starch) is important because of its expansion characteristics, which result from the combined action of fermentation and sun.

Natural fermentation can be replicated by a chemical process, which allows for better quality control, reduced costs and lesser production time. A few enterprises have already adopted the process of adding lactic acid or a mixture of lactic acid and acetic acid to starch in order to suppress the fermentation phase. Nevertheless, these products are always sun-dried, thereby limiting the economic scale of production units.

The emphasis on new products can be seen in the new attitude of Brazilian companies. For example, the largest Brazilian UsUtarch UcUompany, Lorenz, is already producing a totally modified starch, substituting fermentation and sun drying with chemical products by chemical process. The modified starch thus obtained is very successful if used quickly; otherwise, after a period of storage, there are flavour problems and most consumers tend to reject the product. Recently, Lorenz bought the second largest frozen "pão de queijo" producer in Brazil, "São Geraldo", where it is considered, they will probably employ their modified sour starch technology.

As an alternative to the sour starch quality problems, many "pão de queijo" companies are changing their product recipe by eliminating sour starch. Two substitutions are possible: sweet or pre- gelatinized cassava starches, both of which are produced by large modern companies. These products are well established and allow for a stable quality year round, with low microbiologic contamination compatible with international standards. The type of substitution depends on the desired final product. Whereas pre-gelatinized starch is used in dry pre-mixes, and facilitates "pão de queijo" homemade preparation; sweet starch instead, substitutes for sour starch in the refrigerated and frozen "plies de queijo". The Brazilian company, INDEMIL, a "farinha" (flour), sour and sweet starch producer produces dry pre-mix.

Because cassava and cassava transportation represents more than 50% of production costs of sour starch, an important determinant to the level of commercialisation/consumption/total production costs is the price of "pão de queijo". In addition, sour starch production tends to be seasonal, concentrated in the winter months (May to October). In regions like southern Minas Gerais, winter temperatures can be negative and detrimental to cassava production. This leads to price variations (Figure 2) and a greater need to stock sour starch. Finally, seasonality results in companies utilizing less than their full production capacity; thus facing higher fixed (production) costs per unit of product and higher final product prices. This is in contrast to the Centre and North Paraná regions where companies are working between 10 to 12 months a year (Mosquera and Chacon 1992).

The changing face of the industry

To better comprehend the problems of quality and price of sour starch, it is necessary to understand the industrial structure and production process of sour starch.

Sour starch production is a very competitive industry with a large number of small companies (Table 9). The small capacity tends to limit the possibilities to access modern technologies and market information. This explains in part the problems of quality and price instability. Moreover the low educational level of company managers further increases the difficulty of using and accessing modern technologies and market information.

More and larger companies are entering in the market, replacing small traditional units. The managers of the bigger companies typically have better education, with better access to market information and modern technology processes. Bigger production leads to a higher financial capacity, with the possibility to invest in the improvement of product quality and price stabilization. Higher capacities also allow economics of scale, with price decreases. The larger companies are located in the Paraná State. Sour starch units in Central and North-western Paraná also produce sweet starch with large capacity modern units.

Processing technology influences product prices and quality. As noted above, cassava root prices influence sour starch costs, but other costs are also important, mainly salaries (Figure 3).

Sour starch production is very labour intensive, limiting its production by large companies. Sun drying forces units to have large drying surfaces and a large labour force to move the product, limiting the growth of production units and entrance of competition.

Even though modified sour starch production has its limitations, its use will probably increase in the future. Some companies, like sweet starch producers, are making advances in new product lines. Modernization of sour starch production, with more expensive technologies, will make it difficult for small units in Brazil to survive. Lorenz has, alone or in joint ventures, three production units in Paraguay and one in Venezuela, and can produce and export sour starch to Colombia's local market. Moreover the ownership of "São Geraldo" allows Lorenz to dominate most of the product flow process, from raw material to transformation.

Sour starch gives "pão de queijo" a very acid taste, which is appreciated, in Minas Gerais State. In other States however, where consumption is more recent, people do not like the acid taste and prefer instead a softer product. These preferences facilitate the elimination of sour starch and its substitution with sweet or pre-gelatinized starches. Nevertheless, sour starch is still used by most "pão de queijo" producers, in an average mixture of two thirds sweet starch to one third sour starch (Vilpoux et a]. 1995). Sour starch is very useful for its expansion properties, giving the product a better appearance. The advantage to substitute sour with sweet starch is not only quality based, but also price based. A tonne of sour starch costs between US$500 and US$700, compared to a tonne of sweet starch that costs between US$300 and US$400.

Sour starch remains the dominant starch in the market, but this could change in the future. Sour starch production should improve in quality, regularity of supply and price. If it doesn't "pão de queijo" companies will continue to try to substitute sour starch with other kinds of raw material. It will be difficult for the smaller companies to follow the larger companies in these changes. An improvement in sour starch quality and a decrease in production costs will profit companies of all size. Commercialisation of a cheaper sour starch, with better quality will develop markets and provide niches for smaller units.

These problems are difficult to overcome because of the lack of well trained and educated managers and the small size of production units and limited financial resources. A solution is to develop better management and organization within companies, provide managers with better skills and education and to diffuse the industry with market and new technologies information. Market information is necessary for production and price stability. It will also allow smaller companies to have full information on where and for how much to sell their product. Technology diffusion allows quality improvement and production costs to decrease. An opportunity for research is how to facilitate smaller sour starch companies in finding new technologies that will improve product quality. As for starch production, stabilisation of "paes de queijo" prices could improve cooperation between all members of the agribusiness chain, such as final sellers of product, "pão de queijo" producers, sour starch companies and farmers. Circulation of market information must be improved and production planning must be realized between participants in the transformation and commercialisation process.

Cooperation is very important where small companies dominate, as in Santa Catarina State and in the region of Southern Minas Gerais, both in Brazil and other South American Countries, and is also warranted in Centre and Northeastern Paraná where larger units are present. To address this need, an association was created in 1996 in Southern Minas Gerais, and another one in 1998 in Santa Catarina. To date, their activities are still very limited and the organization includes only sour starch companies, but it is a promising start for future developments.

1.2. Cassava as snack foods

Brazil has one main snack food made from cassava called "biscoito de polvilho", or called "pan de yuca" in Colombia. This product is made from sour starch, eggs and water and is baked like bread. Special sour starch properties allow the product to expand like an extruded food, which justifies its classification as a snack food.

The "biscoito de polvilho" industry

This section is based on the research of Vilpoux previously referred to in the "pão de queijo" section. The "biscoito de polvilho" industry, similar to the "pão de queijo" industry, has problems with sour starch quality and price stability. These problems are more serious in the "biscoito" industry, as sour starch is the main ingredient and has no substitutes.

Published data and articles on "biscoito" consumption do not exist. The only information about this product comes from Vilpoux's study of São Paulo State in the second half of 1994 and the beginning of 1995. Because this market appears to be stable Vilpoux's data is still considered relevant. In 1995, 56% of the interviewed managers of supermarkets stated that they believed the market for "biscoito" was stable. Furthermore, 36% of the managers thought that consumption was growing a little but would stabilize in the near future.

Bakeries, supermarkets and shops in gasoline stations sell "biscoito". In bakeries and supermarkets, the product is less frequently found than "pão de queijo". Just over 50% of the bakeries were found selling "biscoitos" but 91% of them are selling homemade product. Similarly, supermarkets do not typically sell "biscoitos" (between 30 and 50%, depending on the size of the shop) (Vilpoux et al. 1995).

Consumption

Children and seniors are the primary consumers of "biscoito". More elaborate products, such as snacks from potato or corn, continue to dominate the snack food market even at higher prices. The 1995 study showed greater growth in the consumption for snacks made from corn and potato than snacks made from cassava.

Consumption of "biscoito" is not very great when compared to "pão de queijo" or other snacks. Very small companies are the sole producers of this product. This is in contrast to the production of corn and potato snacks that are made by a few large companies, the biggest being the Elma chips from the Pescico group. Big companies have national distribution and a known brand.

Constraints in "biscoito" production

The 1995 study showed that one problem of the "biscoito" industry was the lack of marketing, which is particularly important for commercialisation in supermarkets.

Another problem associated with the consumption of cassava-based snack foods is quality of product. "Biscoito de polvilho" is naturally expanded, without extruder. The problems of quality of expansion which depends on the quality of sour starch have already been discussed above. Other quality problems relate to product conception such as:

Colour of the product: yellow in colour makes it appear that the product is dirty.

Flavour of the product: all "biscoitos" have the same flavour, which is given by the sour starch.

Alternative flavours exist, such as bacon and cheese, but such products are very rare.

Bad quality of packaging: all "biscoitos" have similar packaging, made of transparent grey poor quality plastic, with a few inscriptions. The packages seem dated and do not enhance the inside product.

Growth opportunities in the "biscoito" industry

Cassava-based snacks have faced a lot of difficulties, but stable market opportunities are now appearing. These problems are reviewed here before an analysis of the possible solutions for the development of the industry.

Poor product characteristics and absence of variety prejudice cassava-based snacks, which are typically thought of as a poor person's product because of their low selling price. To increase the consumption of "biscoito", it is necessary to change the image of the product. But this change will not be sustainable unless quality control is maintained in the marketplace. The following steps are suggested:

New recipes: develop products with different flavours that will influence customer preferences.

New packaging: develop modern packaging that gives the appearance of a modern product that can be consumed by all kinds of people.

New products: develop a more regular product with an agreeable colour. Experiment with different forms of cassava based snack products that can compete against existing potato and corn snack products.

Improvements of product characteristics through research have been undertaken by technical researchers at CERAT in São Paulo State, CTAA/EMPRAPA in Rio de Janeiro and CIAT in Colombia, in collaboration with market studies to identify consumer's preferences. This is because most cassava-based snack companies are too small to finance research and development. Ideally, "biscoito" producers should be extensively involved to allow better and wider adoption of new products by the industry. Producers think of marketing strategies to increase their sales only after product improvements have been developed.

The problems encountered by "biscoito de polvilho" companies, excluding the one linked with sour starch, are quite different from the problems of other cassava products. "Pão de queijo" and cassava starch are popular products, with great market potential. "Biscoito" on the other hand has a localized market but great potential in the snack food industry. Unfortunately, the product has low quality and is difficult to develop. As earlier highlighted, the first major step to improve the product is to increase quality and improve its form and packaging. Only after such changes, together with renewed marketing strategies, can companies expect market increase.

The Colombian "biscoito" is similar to the one sold in Brazil, competing in the snack food market alongside corn and potato-based products made by very large companies. The problems faced by cassava snack foods in Brazil are similar to the problems faced in other Latin American countries where similar products exist, and the conclusions made for Brazil would appear to be valid for these other countries.

Extruded chips

Another snack food made from cassava is extruded chips, which is sold in Europe and in some Latin American countries. In Europe, these chips are sold in supermarkets, and are very similar to extruded corn products. For example, the TaiYang Company makes a prawn-flavoured product in France using Thai cassava starch.

Similar products exist in Brazil and Colombia under the name Mandiopan and Fritopan, respectively. These products are not ready-to-eat. They have to be fried to allow the expansion of the product. The need to deep-fry the product makes it difficult to market. Perhaps the commercialisation of a ready-to-eat snack in Latin America could be developed as has been done in Europe.

Because information on cassava-based snack products is not readily available, this is an area for future study. More information would allow producers to better define the market potential of these products and what has to be done to achieve success.

1.3. Cassava as convenience foods

The frozen cassava food industry

The growth of the frozen food industry is the result of many factors such as the growth of cities, the increasing distance between work and home and the accelerated pace of human life. Brazilians consume 3.6 million tonnes of frozen food, which is still small when compared to the U.S. market of 14.5 million tonnes (Azevedo 1996). Data from the Brazilian Food Industry Association (ABIA) that frozen and dehydrated foods are the fastest growing segments of the economy in 1997 (Novidades na mdustria de alimentos 1998b).

Five years ago, the Agricultural Cooperative of Cotia (CAC), was the only large enterprise selling frozen cassava products similar to French-fried potatoes. Today there are several frozen cassava and cassava-based products in the market, produced and distributed by several different sized enterprises.

Producing a more convenient cassava product, of good quality, at a competitive price could increase fresh cassava consumption. In general three characteristics are important to consumers of frozen cassava products: appropriate price, ease of preparation or consumption (without peeling, cooking or even frying) and good quality (softness, taste and colour).

Consumption and commercialisation

Data from the State of São Paulo^[3] indicate that per capita cassava consumption is estimated to be 2.5 kg/year with 20% of the production in post-harvest losses. Assuming frozen cassava products are more affordable for higher income groups, the potential demand for such products is estimated at 14 000 tonnes per year.

According to (Azevedo 1996), the frozen food market in Brazil is concentrated in the large urban centres with 80% of the demand driven by the high social classes. Assuming a plant size of 10 tonnes per day, working 25 days per month and 10 months per year, the market for frozen cassava within São Paulo State would require at least five processing plants.

Based on São Paulo State's estimates, the Brazilian frozen cassava demand for the next ten years is between 30 000 and 50 000 tonnes per year. This estimation includes 10 to 15 large Brazilian cities such as Rio de Janeiro, Porto Alegre, Curitiba, Belo Horizonte, Fortaleza and Salvador. Initially, these potential markets may be restricted to only the largest cities since small informal firms close to cassava producing areas may supply the small cities.

In Brazilian supermarkets, the consumer price for frozen French-fried potatoes is US$3.37/kg compared to the consumer price for fresh potatoes of US$0.75/kg. A similar difference is encountered between frozen cassava and fresh cassava with consumer prices of US$2.36/kg and US$0.78/kg, respectively. As these two products are easy substitutes, the cheaper price for the frozen cassava product could theoretically allow for greater consumption.

A cassava producer's cooperative established a processing plant in 1997 for the production of frozen cassava products with a transformation capacity of six tonnes of roots/day. Table 10 evidences that cooperative prices are 61% higher than those paid by middlemen in traditional fresh markets.

The Table also shows that the frozen cassava consumer is paying 66 percent more than the fresh cassava consumer price, with 18% of the difference corresponding to processing costs. By comparison, the supermarket margin is greater for traditional cassava than for frozen cassava. Within the frozen cassava market structure, middlemen are eliminated but processing costs are high.

Individual quick frozen process and the cassava industry

The CONGELAGRO enterprise in Colombia produces the cassava-based product called "croquette" with a consumer price of almost US$2. This enterprise uses the individual quick frozen (IQF) process that substantially increases the processing costs. Although they have some problems in raw material supply, in general their product is well positioned in the market.

The process developed by Cereda and colleagues, uses a pricking operation before freezing the product. This is a process that requires an intense use of hand labour since specialized machines for cassava peeling or cutting are very expensive. The fresh roots first undergo a washing pre-peeling treatment in a tumbling washer. During this treatment, the dark outer skin of the root is removed. In the next step, the roots are cut in cylindrical pieces and then the thick outside fibrous layer -the cortex - is removed by hand. After cutting, the cylindrical pieces are washed again before transforming them into the final shape. This shape can be semi-cylindrical or toothpick cassava. A selection of the toothpick cassava will define a second quality cassava that will be used after cooking for elaboration of cassava "mass". Afterwards, the product is pre-cooked in a boiling solution with water, salt and monosodium glutamate. The product is then cooled by submerging it in a cooler tank and placed in trays that will go for a pre-frozen period in the container. After this period, the toothpick cassava and the semi-cylindrical pieces are easily handled and packaged (Cereda et al. 1997).

This process was installed by the CAU cooperative at an initial investment of about US$31 000 including a rented container and a free building provided by the municipality. Without municipal support, it was estimated that the investment for an eight tonnes per day procession plant would be approximately US$174 000 (US$130 000 for a 300 square meter building space with pre-frozen and frozen rooms plus US$44 000 for equipment). A recently installed plant in the State of Paraná had a three hundred thousand dollar investment in a new cassava product. This plant uses a French cooking system with steam and high pressure and temperature with an actual capacity of 120 tonnes per month and predicted potential for the second year of 210 tonnes per month (VAPZA lanca mandioca pronta 1997a).

Constraints in the frozen cassava industry

High initial investment costs are one of the difficulties encountered by farmers trying to enter the frozen cassava industry. Another constraint facing the frozen cassava industry is the consumer preference for fresh cassava. Whilst some consumers are aware of the cooking quality of frozen cassava, others still believe that frozen foods loose their quality characteristics. Secondly frozen cassava consumption is a new and developing market. Other difficulties emerge in the product distribution and marketing activities because of high transportation costs and the lack of frozen capacity in many markets.

Co-operatives benefit from vertical integration where raw material supplies, planting and harvesting activities can be programmed. On the other hand, the distribution and selling of product cannot be efficiently approached by cassava farmer cooperatives. It is also difficult for these cooperatives to reduce processing costs because of the high cost of new technology and equipment. In contrast larger companies may face several challenges with raw material supply but they can make large investments in processing technologies that reduce distribution and selling costs dramatically.

Over time, the frozen cassava market has experienced firms constantly leaving and entering. Firms motivated by the increasing demand for frozen foods enter. Firms leave because of a lack of raw material supply and the high costs of processing, distribution and selling. Three types of firms can be identified in this scenario:

1) Large private firms selling cassava and other frozen products,
2) Small private firms working in medium-city markets and
3) Cooperatives of cassava producers.

In the future large firms will try to dominate the market through strategic alliances with cassava suppliers and processors. The CAU cooperative signed a contract, selling 80% of its production to a large frozen products firm, the Pratigel. Small private firms sometimes work informally and will be restricted to a small piece of the market. Cooperatives of cassava farmers will need support from government or private agencies in order to reduce processing costs and to develop technologies and new cassava-based products. In general, promotion will require research and development (R&D) on: the identification of appropriate cassava varieties for human consumption; on adequate technology for cassava processing including peeling, cutting, frying, frozen and storage; on identifying market opportunities; and on identifying areas for cassava production with experienced farmers.

Four or five stakeholders (CAC, CAU, PRATIGEL, and VAPZA) can be identified within the Sào Paulo State market and even if the frozen cassava products of each one differ, they are facing difficulties in the market because of high processing and distribution costs.

It will also be difficult to achieve market expansion because the consumer price is still very high and the target consumer group corresponds to high-income classes. The challenge for these enterprises will be to look for cost reducing improvements within the process. The strategy for market expansion is a reduced consumer price for frozen cassava.

Frozen cassava processing has to be improved in order to reduce production costs and consequently reduce the final price to consumers. As the demand for frozen cassava increases in the future, the key factors for enterprises remaining or entering in the market will be raw material supplies, low processing cost, high processing efficiency and quality and new cassava products development. As large multinational firms, such as Bonduelle, McCain and Fri-D'or, enter the frozen vegetable products market, improved marketing strategies and publicity of frozen foods could positively benefit the expansion of the frozen cassava market.

Future directions

The domestic market for fresh cassava, cassava snack and convenience foods is often overlooked. These markets exist as growth markets in part because of growing urbanization, and in part because cassava-based products can be competitively priced with alternative foods.

The global analysis indicated that the increased demand by 2005 for cassava as a food, owing to urban expansion, could be as much as 16 percent of 1995 production in Africa, 6 percent in Latin America and 8 percent in Asia. The analysis suggested that nine African, two Asian and two Latin American countries could experience a growth in demand of more than twenty percent of production by 2005 owing to urbanization. An additional 26 countries could experience an increase in demand equivalent to 10 percent of production. For this potential to be realized, it is necessary that the food products must be delivered to the urban centres in a form that is appealing, convenient and price competitive.

The Brazilian example of "pão de queijo" has demonstrated that cassava based products can enter the mainstream fast food market. The case study also demonstrates that as the market expands, issues of quality, ease of use and timeliness become more important. In this case, growth is beneficial to cassava producers, but threatens the existence of small-scale processors. If the processors are to continue to exist, they need to find and adopt new technologies that would ensure that their product is of an acceptable and reliable quality. To achieve this end, the processors may require assistance in developing new technologies, and may further require training to enable them to work effectively in a growing and larger market.