Issue 7, September 2011

Biocontrol offers benefits to Africa
From traditional to science based
Amazing maize
Local seeds and social networks
What sustains productivity?
Cassava processing research
Impact of R4D on farmers
Funding agricultural R&D
Outcome mapping
IITA’s new social science research agenda

Download PDF

The benefits of social science

Woman peeling cassava. Photo by IITA.
Woman peeling cassava. Photo by IITA.

Agricultural research is the key to achieving IITA’s mission of enhancing food security and reducing poverty in sub-Saharan Africa. Hence, IITA undertakes research with and for the people and engages a whole range of partners along the research-to-development continuum. The effectiveness of this approach depends, however, on the richness of the social science context that is required to ensure the relevance of agricultural research in the discovery, adaptation, adoption, and diffusion of new technologies and institutional innovations.

The focus and methods of socioeconomics research, particularly of impact assessment, have evolved over time in response to donors’ interests and research mandates.

Traditionally social sciences was narrowly defined and focused on working in collaboration with biophysical scientists on issues related to technology generation and delivery. The research agenda centered on several sets of key questions: the extent of and constraints to adoption; the impacts of technology adoption on yields and household incomes; and ex-ante (or expected) benefits from new technologies. From a rather narrow emphasis on the adoption of new varieties in the 1970s, the focus has now expanded to estimating rates of return to research investments in the 1980s and to examining a wider range of impacts and the distribution of benefits across different socioeconomic groups after the 1990s.

Currently, IITA’s socioeconomists undertake a wide range of socioeconomic and impact assessment activities supporting broader technology development and delivery efforts.

This issue presents highlights of some recent research in socioeconomics and impact assessment and IITA’s social science research agenda for the next decade. A set of studies addresses strategic, macro-level impact and policy issues and offers strategic information and analyses. For instance, one study showed significant productivity gains realized after the mid-1980s, driven principally by agricultural research and development (R&D), improved weather, and policy reforms.

Another study found that, with the successful implementation of emerging national strategies for the agricultural sector, agricultural growth is expected to increase from 4.6% under a business-as-usual scenario to 6.4% with the implementation of national strategies.

One study exploring technological and policy options for forest and biodiversity conservation in West Africa showed that strategies to reduce deforestation and conserve biodiversity must focus on transforming agricultural practices from traditional to modern science-based methods.

Several other recent studies also address the extent, determinants, and impacts of adoption of a range of production and processing technologies and institutional innovations developed and promoted by IITA and partners.

Social science is a vital dimension to our biological science.
— DG Hartmann

Amazing maize: Investment in agricultural research pays off

Maize is a major food staple in sub-Saharan Africa. Photo by IITA.
Maize is a major food staple in sub-Saharan Africa. Photo by IITA.

Researchers have shown that investment in maize research in West and Central Africa pays off. A study by IITA’s agricultural economists reveals that the generation and diffusion of modern maize varieties in the last three decades have lifted more than one million people in sub-Saharan Africa out of poverty.

Over half of this impact can be attributed to international maize research at IITA and the International Maize and Wheat Improvement Center (CIMMYT).

This was reported in a brief by the Standing Panel on Impact Assessment of the Independent Science and Partnership Council. The brief was based on the paper by Alene et al. (2009), who estimated the economic and poverty reduction impacts of international maize research in West and Central Africa from 1971 to 2005.

Based on data obtained from IITA’s financial reports and the FAOSTAT database, a total of US$308 million was invested in maize research between 1971 and 2005, with international maize research accounting for about 66% ($204 million).

Maize research in West and Central Africa had been conducted by IITA, CIMMYT, and partners that include the national agricultural research systems (NARS). IITA, which has had a regional mandate for maize improvement since 1980, started maize research around 1970. With its partners, IITA had developed high-yielding varieties with increased tolerance for multiple biotic and abiotic stresses. As a result, these varieties have contributed to changing the status of maize from a minor crop to one of the most significant food and cash crops in the region.

Breeding programs at IITA and CIMMYT have been the major sources of germplasm for the released varieties, supplying 90% of the germplasm in the 1970s, 60% in the 1980s and 1990s, and 85% since the late 1990s. IITA currently supplies nearly 70% of the germplasm in the region, with little or no further improvement before the release.

In Nigeria, which accounts for nearly half the maize area in the region, 60% of the maize areas were planted to modern varieties in 2005. The total maize area tripled from 2.6 million ha in 1981 to more than 7 million ha in 2005, with the area under modern varieties rising sharply from 111,000 ha to 4.2 million ha. Adoption figures estimated from the early 1980s onwards suggest a steady growth in adoption in the region.

Why maize?
First, maize grows in a wide range of production environments, making it an important source of home-produced food. Secondly, it is a desirable cash crop, providing farmers with income and keeping market processes affordable for the urban poor. Thirdly, resource-poor farmers are able to adopt the predominantly open-pollinated modern varieties without having to buy fresh seeds each season.

Maize processing. Photo by SP-IPM.
Maize processing. Photo by SP-IPM.

Based on the germplasm, international maize research moved between 300,000 and 500,000 out of poverty each year. It is estimated that every $1 million invested in maize research by IITA lifted between 35,000 and 50,000 people out of poverty. With the involvement of NARS in maize improvement work, national programs have also significantly contributed to poverty reduction efforts in the region, with over $100 million being invested since 1970.

Thus, the total net benefit from international and national maize research in the region for 1981–2005 is estimated at $6.8 billion, equivalent to 12% of the present value of total maize production over the same period. Annual net benefits increased from $43 million in 1981 to >$400 million in 2005, with an annual average of $274 million (in 2000 constant prices).

Maize improvement research in the region had a benefit-cost ratio of 21. This means that every dollar invested in maize research generated additional food worth $21. Estimates for country-level benefit-cost ratio ranged from 11 in Mali to 84 in Nigeria, with an average rate of return of 43% in West and Central Africa.

Bottom line
Maize research has generated a stream of benefits in the region, and is thus considered a worthwhile investment. This underlines the importance of international research.

Study results also suggest that poverty in the region could have been much worse had there been no research and improvement in maize yields, when pest and disease pressure, the decline in soil fertility, and expansion into marginal lands are considered.

Research on nonyield benefits, such as drought-tolerant maize and varieties for better nutrition, for example, may even show greater benefits, according to the study.

Maize research will continue to be a powerful factor in reducing poverty, according to the brief. However, impacts of research investments are conditioned by farmers’ access to inputs, such as fertilizer, credit, seeds, extension and input-supply systems, and market infrastructure.

Alene AD, A Menkir, S Ajala, B Badu-Appraku, A Olanrewaju, V Manyong, and A Ndiaye. 2009. The economic and poverty impacts of maize research in West and Central Africa. Agricultural Economics 40: 535–550.

CGIAR. 2010. Improved maize benefits millions of Africa’s poor. Standing Panel on Impact Assessment, Independent Science and Partnership Council. CGIAR, Washington D.C., USA. 4 pp.

What sustains the productivity of African agriculture?

A study carried out to measure productivity trends and the effects of research and development (R&D) in African agriculture shows that, over the period 1970–2004, African agricultural productivity grew at an annual rate of 1.8%. Agricultural R&D, improved weather, and policy reforms were found to be the principal drivers of the productivity gains realized after the mid-1980s.

Researcher in Bioscience Lab, IITA. Photo by IITA.
Researcher in Bioscience Lab, IITA. Photo by IITA.

The study by Arega Alene, IITA’s agricultural economist, published in Agricultural Economics, showed that investments in agricultural R&D had an annual rate of return of 33%, proof that agricultural R&D in Africa is a socially profitable investment.

The study found that a strong growth of agricultural R&D investment of about 2%/year in the 1970s led to faster productivity growth after the mid-1980s, but stagnation of R&D investments in the 1980s and early 1990s led to slower growth in productivity in the 2000s.

Agriculture is key
Growth in agricultural productivity has been cited as the key to economic growth, and many researchers have in fact looked at the trends and sources of growth in agricultural productivity in developing countries. The extent of recovery of African agricultural productivity since the mid-1980s, however, varies widely, depending on the methods used to measure and explain it.

The study looked at total factor productivity (TFP) growth in African agriculture using available data on all African countries for the period 1970–2004. Data on agricultural production and conventional agricultural inputs for 52 African countries for the period 1970–2004 were obtained from the FAOSTAT database (FAO 2007). Meanwhile, data on agricultural research investments for 15 African countries for 1971–2001 were obtained from the Agricultural Science and Technology Indicators database of the International Food Policy Research Institute (IFPRI). These countries were Bénin, Botswana, Burkina Faso, Côte d’Ivoire, Ethiopia, Ghana, Kenya, Madagascar, Malawi, Mali, Niger, Nigeria, Senegal, South Africa, and Zambia.

Soybean field. Photo by IITA.
Soybean field. Photo by IITA.

Using conventional indices of productivity growth, the study estimated that the annual aggregate productivity growth in African agriculture was only 0.3% over the period 1970–2004. The poor aggregate performance was due to a decline in agricultural productivity in over one-third of the sub-Saharan African countries. With an annual growth rate of only 0.1%, the conventional approach implied that the performance of agriculture in the region was poor and that agricultural productivity stagnated.

In sharp contrast, the improved measures of productivity showed that African agricultural productivity grew at a much higher annual rate of 1.8% over the period 1970–2004. In sub-Saharan Africa, agricultural productivity grew at an annual rate of 1.6% over the same period. As expected, North African countries experienced a higher annual productivity growth rate of 3.6%. Although 20 countries experienced annual productivity growth rates of over 2%, only seven countries (Burundi, Comoros, Democratic Republic of Congo, Côte d’Ivoire, Lesotho, Mozambique, and Sao Tome and Principe) experienced negative productivity growth rates, due largely to declining technical efficiency.

Technology drives agricultural productivity
Productivity decline during the 1970s was attributed to technological regress (−1.1%/year). However, technical progress (1.5%/year) was pinpointed to be the principal source of recovery of productivity during the 1980s.

The new measures demonstrated positive annual productivity growth in all three periods: 1970s (1.4%), 1980s (1.7%), and during 1991–2004 (2.1%). Unlike the conventional estimates, the improved measures demonstrated sustained increases in productivity growth over the years, with an impressive annual growth rate of over 2% achieved during and after the 1990s.

Motorized cassava grating machine commissioned by IITA-CFC in Sierra Leone. Photo by IITA.
Motorized cassava grating machine commissioned by IITA-CFC in Sierra Leone. Photo by IITA.

The results further showed that rainfall is positively and significantly related to agricultural productivity. This confirms that the weather is a critical constraint to agricultural production in Africa.

Despite the fluctuations in productivity induced by weather fluctuations, both trade and agricultural productivity exhibited an increasing trend after the mid-1980s.

The results showed a positive and significant association between trade policy reforms and productivity in African agriculture, suggesting that policy reforms indeed contributed to the recovery of agricultural productivity after the mid-1980s.

In particular, agricultural productivity grew at an impressive annual rate of over 2% after the early 1990s. This is consistent with recent data on economic recovery in Africa, as shown by stronger growth rates in agricultural gross domestic product (GDP) following improved macroeconomic conditions and commodity prices after the mid-1980s. The results demonstrated that technical progress, rather than efficiency change, was the principal source of productivity growth in African agriculture.

Alene said that a 10% increase in R&D investments would raise agricultural productivity by 2%/year. With an annual rate of return of 33%, R&D has proved to be a socially profitable investment in African agriculture. The analysis points to the need for increased investments in agricultural research to sustain productivity growth in African agriculture.

Alene, Arega D. 2010. Productivity growth and the effects of R&D in African agriculture. Agricultural Economics 41: 223–238.

Local seeds and social networks

Including seeds of local crop varieties in the relief seed packages distributed to small-scale farmers after natural calamities could help indigenous crop diversity to recover faster. In addition, existing social networks which act as vital channels for seed distribution hasten the recovery of diversity in disaster-affected communities. These are among the findings of a recent study by IITA that looked into the loss and subsequent recovery of cowpea diversity in Mozambique when widespread flooding, followed by severe drought, hit most of the country about 11 years ago.

Diversity of cowpea seeds. Photo by IITA.
Diversity of cowpea seeds. Photo by IITA.

Farmers in Mozambique usually receive relief seed packages as a stop-gap measure to alleviate the effects of natural disasters that often wipe out their crops. However, most of the seeds are generally of introduced and genetically uniform varieties purchased from markets or provided by seed companies or by well-meaning relief agencies, which slow the recovery of crop diversity.

The study noted that the speedy recovery of Mozambican cowpea diversity after the back-to-back disasters of 2000 was largely due to the exchange of seeds among farmers through making gifts and other social interactions involving friends, family members, and relatives within the same community or those adjacent to it. Morag Ferguson, a molecular biologist with IITA and one of the study’s lead researchers, says that farmers in Africa traditionally grow many crops and several varieties of each crop on the same plot of land to cope with unforeseen economic or environmental instabilities. They usually set aside part of their harvest to serve as seeds for the next cropping season. They also share or trade some of these seeds with friends and relatives. When natural disasters strike, many farmers often lose the seeds that they have set aside and are forced to rely on relief seeds, buy seeds from the market, or receive seeds as gifts from friends and relatives.

“We found that the substantial recovery of cowpea genetic diversity two years after the calamities was mainly due to the informal exchange of seeds among farmers that served as a socially based backup for the safety of crop diversity. It is therefore important that seed relief strategies recognize and capitalize on this existing traditional network, based on social relations, to help restore diversity especially after natural upheavals,” Ferguson said.

The study was initiated in 2002, two years after the floods-then-drought disaster, in Chokwe and Xai Xai districts of the Limpopo River Valley—areas that were among those severely affected. The findings of the research have been published in the current issue of Disasters, a publication of the Overseas Development Institute.

The research established that nearly 90% of the farmers in the affected areas received cowpea relief seeds immediately after the back-to-back calamities. Two years afterwards, only one in every five of the recipient farmers were still growing the seeds, whereas more than half sourced their seeds from markets. However, this did little in restoring cowpea diversity in the affected communities as the seeds bought by farmers from the market were mostly uniform, since they came from other districts that grew just one variety or a few select varieties.

Social networks provide a safety net for people affected by disasters. Photo by IITA.
Social networks provide a safety net for people affected by disasters. Photo by IITA.

On the other hand, about one-third of the affected farmers obtained seeds from friends and relatives from nearby districts not affected by the disaster and with excess seed to restock their farms—the same people with whom they had been exchanging seeds before the disasters. This practice was the main reason why cowpea diversity was restored in these areas, the study showed.

Ferguson says that such a seed distribution system based on social relations is already in use in an approach developed and implemented by the Catholic Relief Services in partnership with other relief agencies in which seed vouchers are exchanged for seeds at ”Seed Fairs”. In this approach, farmers from nearby districts not affected by disaster and with surplus seeds come to the Seed Fair to sell seeds to disaster-affected farmers in exchange for vouchers, which they then cash-in with the relief agency.

“This approach recognizes that farmers’ seed systems are robust and resilient, and can provide seeds even in emergencies. This study shows that such an approach will be more effective in restoring diversity faster and more efficiently than a system based on direct distribution only.”

The study was the first of its kind to investigate in detail the effects of disasters on crop diversity and its recovery. It combined agronomic observations (e.g., looking at the seeds’ color, size, pattern, and shape) with biotechnology tools to determine the seeds’ genetic makeup.

From traditional to science based: Transforming agricultural practices

In recent times discussions on deforestation in the tropics more often than not have pointed to agricultural expansion as one major factor behind the depletion of forests.

Forests are crucial to life on earth: IITA forest. Photo by IITA.
Forests are crucial to life on earth: IITA forest. Photo by IITA.

This argument has been underpinned by the fact that agricultural growth in the region has been driven by area expansion rather than improved productivity.

Environmentalists say the depletion of forests hurts biodiversity, encourages climate change, and jeopardizes our future existence on this planet.

But a new study finds that increasing agricultural productivity through the application of fertilizers will reduce the rate of deforestation and help transform agriculture with less damage to the environment.

The study by researchers Jim Gockowski of IITA and D. Sonwa of CIFOR, two centers of the CGIAR, established that the boom in production in the last two decades in the major cocoa-producing countries of Côte d’Ivoire, Ghana, Nigeria, and Cameroon was detrimental to the forest, as farmers had to clear large expanses of trees for cocoa cultivation.

Cocoa production, they say, doubled between 1987 and 2007 but at a heavy cost, as West Africa’s Guinean Rainforest (GRF)—a region described as the ‘global biodiversity hotspot’—shrank to 113,000 km2.

The principal driver of this environmental change has been the expansion of low-input smallholder agriculture that depends on environmentally destructive practices, such as slash-and-burn and land clearing.

The researchers found that increasing the use of fertilizer on cocoa–timber farms would have spared about 2 million ha of tropical forest from being cleared or severely degraded.

Cocoa farmer drying beans. Photo by IITA.
Cocoa farmer drying beans. Photo by IITA.

The study suggested that farmers could have achieved the same outputs without widespread deforestation through the intensified use of fertilizers and agrochemicals coupled with improved crop husbandry.

By doing so farmers would have doubled their incomes and helped to avoid deforestation and degradation. This would have generated a value of over US$1,600 million on 1.3 billion tons of CO2 emissions that would not have come as a result of the deforestation.

The findings should be taken into consideration in discussions about efforts to reduce emissions from deforestation and degradation (REDD), say the researchers. Instead of considering complicated strategies involving monetary or in-kind transfers to farmers or communities for altering their land- use behavior, funds to support REDD could be used to provide incentives and promote agricultural intensification efforts that would lead to higher rural incomes, greater food security, and avoid emissions through the achievement of higher agricultural yields.

The limited use of fertilizer in the GRF (less than 4 kg/ha of total nutrients) may have been logical in 1960, when West African populations were only 25% of today’s levels and forest land was still relatively abundant. That choice is no longer tenable in a context where only 15 to 20% of the GRF remain. There are no longer any frontier forests in West Africa for future generations to exploit.

Strategies to reduce deforestation and conserve biodiversity in West Africa must thus focus on transforming agricultural practices from the traditional to modern science-based methods. Fertilizers- for-Forest (F4F) technology is available to sustainably intensify production and has achieved impressive increases in cocoa yield on a limited scale in parts of the GRF.

The authors say that REDD funding support to mitigate climate change as discussed in the Copenhagen Accord offers the potential of significant new public resources for investments in agricultural research and extension and market infrastructure to support the transformation of traditional agriculture in West Africa. The estimated value of the CO2 emissions thus avoided is conservatively estimated at $565/ha for achieving the envisaged doubling of yields. A significant proportion of REDD+ funding should be used to increase the adoption and level of fertilizer use in an F4F program.

Cassava processing research in Nigeria

Women making gari. Photo by IITA.
Women making gari. Photo by IITA.

Between 2002 and 2010, IITA implemented the Integrated Cassava Project (ICP) to support the Presidential Initiative on Cassava. Under ICP, IITA and its partners successfully introduced and promoted more than 40 cassava varieties to Nigerian farmers, and facilitated the establishment of hundreds of processing centers and fabricating enterprises.

The Presidential Initiative was launched in July 2002. It aimed to create awareness among farmers on the opportunities in the cassava markets worldwide, increase the crop’s area of cultivation to 5 million ha targeting a harvest of 150 million t annually, and earn Nigeria up to US$5 billion every year from export, by the end of 2010.

It also targeted to produce 37.5 million t of processed cassava products annually, such as gari, pellets, chips, starch, and ethanol for local and export markets.

The Nigerian government provided IITA some funds to assist in its R4D efforts and implement the ICP. The project had two components: the Cassava Mosaic Disease (CMD) Preemptive Project which focused on production aspects through breeding and distribution of improved varieties resistant to the virulent Uganda variant of CMD, and the Cassava Enterprise Development Project (CEDP). This promoted the development of enterprises associated with cassava processing.

Through the project, IITA successfully introduced and promoted new varieties to farmers via the national agricultural research system (NARS), especially the Agriculture Development Program (ADPs). It also facilitated the establishment of many processing centers and fabricating enterprises between 2002 and 2010, contributing greatly to the development of the Nigerian cassava industry.

The study
A study was carried out to look at the impact of IITA’s processing research on Nigeria’s staple food system and to draw lessons from these interventions.
It addressed the following research questions:
• What is the level of awareness and adoption of improved cassava varieties, and other production and processing technologies?
• What are the effects of adoption of these production and processing technologies on farming households, their villages, the fabricators, and processors?

The survey was carried out in 70 villages in the four geopolitical zones in Nigeria: South-West, South-East, South-South, and North Central where 952 farmers, 143 processors, and 58 fabricators were interviewed.

Partial budgeting methods as well as micro-econometric evaluation methods were used to assess causal effects based on the changes in outcome and impact indicators in areas with the cassava processing research interventions relative to those without the interventions.

Sorting cassava. Photo by IITA.
Sorting cassava. Photo by IITA.

Technology adoption and benefits: Village level
The results showed that from 2002, the area of land under cassava production had increased by 17% in intervention villages and by 10% in non-intervention villages. Also, the crop was found to occupy more than 70% of total area available for food crops in the sample villages.

Processing machines, such as graters, pressers, fryers, grinders, dryers, and millers proliferated in intervention villages; small percentages of other processing machines, such as boilers and fermentators, were also found.

Cassava produced by farming households was consumed or sold, fresh or processed, with some going to waste. There was a decrease in the proportion of waste over the years in both villages; the same was observed in the sale of fresh cassava roots. On the other hand, the volume of processed cassava increased over the years, suggesting the positive influence among cassava farmers of the government’s efforts to boost production and processing.

Gari and fufu remain the most popular cassava products as they were a decade ago. However, in all villages especially in those where interventions were introduced, odorless fufu, starch, and chips were becoming increasingly popular. Other products such as cassava flour and ethanol were found but in small quantities and only in the intervention villages.

Technology adoption and benefits: Farming households
Cassava was the most important crop grown followed by yam, maize, and plantain/banana, among others. It occupies about 43% of the total cropland in the study areas.

The adoption rate for improved cassava varieties varied among farming households: 74% in intervention villages going up to 94% among those that had attended R4D training and 65% for the other locations.

On the adoption of processing machines, the grater was the most important with 60% adoption in intervention communities and 76% among those that had attended R4D training. It was followed by the presser; the adoption rate for other processing machines was found to be generally low.

Results also indicated that the adoption rate of improved cultivars was significantly greater in intervention villages than in nonintervention villages, with the use of graters for processing cassava having a positive and significant influence on adoption of new varieties. Factors that influenced the intensity of adoption of graters included contact with extension agents and the use of improved varieties, among others. Although adoption of varieties and the uptake of graters reinforce each other, the effect of improved varieties on adoption of graters was found to be stronger.

Given the greater influence of adoption of improved varieties on adoption of processing machines, a sequential approach should be used in technology delivery involving improved varieties and processing technologies. The sustainability of cultivation of improved varieties is ensured by the availability and use of cassava processing machines among the households in the villages.

Adoption of production and processing technologies took place mainly in the last 20 years, particularly in the last decade; 62% of respondents indicated that they started using improved varieties in the last 10 years, whereas 43% had started using the grater and 45% the presser. Responses were similar for other processing machines. This was the period when IITA and collaborators had intensified the push for production and processing technologies in Nigeria prompted by the Presidential Initiative.

The number of households processing cassava into various products had increased by 21% compared with 10 years ago. Gari (57%) and fufu (30%) were the most popular products. The remaining 13% was shared by other products, including cassava flour and starch, among others.

The gross margin values and the benefit-cost (B:C) ratios were greater for improved cassava ($4090) per hectare than for local varieties ($1500). The B:C ratio was 3:9 in favor of improved cassava varieties with the difference being attributed to the relatively high adoption of both improved cassava varieties and various management techniques extended to and used by the intervention villages.

Grinding cassava. Photo by IITA.
Grinding cassava. Photo by IITA.

Cassava machine fabricating enterprises
The study showed that 66% of the enterprises fabricating machines were small scale and that 79% of them were owned by sole proprietors. There were varying levels of awareness on the different types of processing machines with many fabricators using other machines not meant for cassava processing. Graters (85%) and pressers (83%) were the most popular machines fabricated, followed by grinders/millers (59%) and fryers (41%).

IITA-contacted enterprises performed better than the other fabricators. Factors influencing the numbers of machines produced included being a sole proprietor, year of establishment (2003–2010), experience in manufacturing machines apart from cassava-based types, availability of spare parts, contact with IITA/collaborators, and revenue from selling the machines. However, being a sole proprietor and the availability of spare parts had the greatest influence on production.

Processing enterprises
Half of the enterprises interviewed said that they had had contact with IITA/collaborators and 41% had participated in R4D training on cassava processing in Nigeria. The processing machines mainly used by these processors included graters (85%), pressers (67%), fryers (64%), and sifters (34%).

Adoption of cassava products, such as gari, starch, bread, high quality cassava flour (HQCF), instant fufu, odorless fufu, and broiler meal was generally low but still higher for those processors that came into contact with IITA than for those that did not. Adoption was higher for gari (70%), starch (14%), HQCF (16%), and odorless fufu (17%).

An analysis of gari processing showed that it costs less to use “machines only” for processing compared with “manual and machines” and “manual only”. From the cost analysis, $92 was saved when using machines to process 1 t of fresh cassava into gari compared with manual processing.

In conclusion, results indicate that gari is still the most popular cassava product (80% of households and 70% by processing enterprises). Graters and pressers were the most widely used equipment in Nigeria. Although the study found some reduction in adoption due to the decline in the implementation of the Presidential Initiative, most processors and equipment fabricators were still operating in the area. Despite a slow growth, processing increased with a high adoption rate of improved varieties (68%) by farmers. Processing equipment, such as graters and pressers, are now being used widely.

Biocontrol offers benefits to Africa

Biological control programs implemented by IITA and partners on cassava green mite have brought benefits worth more than $1.7 billion to Nigeria, Bénin, and Ghana in the last 18 years.

Diseased plant. Photo by IITA.
Diseased plant. Photo by IITA.

Ousmane Coulibaly, IITA Agricultural Economist, describes the figure as “a conservative estimate.”

“The figure represents the amount those countries would have spent over the years on other methods such as chemical control and/or yield losses if they never adopted biological control,” said Coulibaly.

The cassava green mite is a pest that was responsible for a yield loss in cassava in Africa of between 30 and 50% until a natural enemy of the pest helped to contain the devastation. In 1993, scientists from IITA and partners identified Typhlodromalus aripo as one of the most efficient enemies against cassava green mite. The introduction of T. aripo reduced pest populations by as much as 90% in the dry season when pest populations are usually high; in the wet season, pest attacks are not as severe.

T. aripo from Brazil was first released on cassava farms in Bénin and, subsequently, in 11 countries; it is now confirmed as established in all of them, except Zambia. T. aripo has also spread into Togo and Côte d’Ivoire from neighboring countries. It spread to about 12 km in the first year, and as much as 200 km in the second year. Today, the predator of the cassava green mite has been established on more than 400,000 km2 of Africa’s cassava-growing areas.

Scientists say chemical control of the pest was ruled out because of possible adverse effects of chemicals on illiterate farmers and the environment. Also, disease pathogens and pests tend to develop gradual resistance to chemical pesticides over time. Moreover, most chemical pesticides are not selective and might destroy the natural enemies and the pests together.

Coulibaly notes that since the release of T. aripo, benefits in Nigeria have been estimated at $1.367 billion, followed by Ghana $305 million, and Bénin $54 million. Consumed by more than 200 million people in sub-Saharan Africa, cassava is a staple food that is rich in calories, highly drought tolerant, thriving in poor soils, and easy to store in the ground.

Impact of cassava R4D on smallholder farmers

Does research-for-development (R4D) have an impact on small-scale farmers? The answer is a resounding ‘yes’ based on a series of impact studies on IITA’s cassava R4D work in the Democratic Republic of Congo (DRC) and Malawi.

Cassava harvest in Bungu, Tanzania. Photo by E. Kanju, IITA.
Cassava harvest. Photo by E. Kanju, IITA.

Democratic Republic of Congo
In DRC, an emergency response R4D program was implemented from 2001 to 2009 after the outbreak of the cassava mosaic disease. The study provided hard evidence of a clear impact on household access and participation in markets, adoption of improved crop varieties and crop management practices, plot yields, gross margins, and food security.

Cassava is the number one provider of staple food and wages in DRC, accounting for more than 70% of the annual crop area and supplying around 56% of the calories in the diet (FAO 2010). In 1996, a new, more virulent Ugandan strain of the East African Cassava Mosaic Virus (EACMV-Ug) was detected. By 2000, it had spread to most cassava-producing regions. Most of the widely planted varieties had no resistance to EACMV-Ug and it was feared that the disease would lead to widespread crop losses and food insecurity.

The emergency response program to the outbreak was started in 2001 and aimed at increasing farmers’ income, improving food security, and nutrition, and reducing poverty. It was supported through a multi-donor funding basket.

The first phase was implemented from 2001 to 2006 in the western provinces of Bas Congo, Kinshasa, and Bandundu because there was war in the eastern part of the country. The second phase, from 2007 to 2009, was expanded to include the central and eastern provinces: Equateur, Province Orientale, Katanga, Kasai Oriental, Kasai Occidental, Maniema, Nord-Kivu, and Sud-Kivu.

The first phase focused on the rehabilitation of cassava production through the multiplication and distribution of clean planting material of existing, already released cassava varieties, breeding for improved varieties with resistance to the viral disease and acceptable consumer traits, and improved crop management technologies. The second phase added components of postharvest management.

The program was implemented through an agricultural R4D approach which brought together different R4D organizations into public–private partnerships with clearly defined roles. These included the Programme National Manioc (PRONAM) within the Institut National pour l’Etude et la Recherche Agronomiques (INERA), IITA, the South-East Consortium for International Development (SECID), FAO, Centre d’Appui pour le Développement Integral de Mbankana (CADIM), PACT Congo, community based organizations, farmers’ associations, and village-level farmers’ groups.

Chikwangues, made from fermented and pounded cassava, for sale at Kolo market, Bas-Congo, DRC. Photo by IITA.
Chikwangues, made from fermented and pounded cassava, for sale at Kolo market, Bas-Congo, DRC. Photo by IITA.

The impact
In-depth interviews with participants revealed four stages by which the program interventions had an impact at the farm household level.

First stage: IITA and INERA undertook the multiplication of breeder and foundation planting materials and the development of new varieties, crop management, integrated pest management, and processing technologies. SECID, FAO, and CADIM implemented rapid multiplication and the large-scale distribution of disease-free planting materials, using a quality control system of primary and secondary nurseries to ensure that large quantities of planting materials were supplied to farmers for establishing their crops.
The Bureau Central de Coordination (BECECO), a government clean seed multiplication and distribution program funded by the World Bank, supported the multiplication and distribution of planting materials and farmers’ training. Community based organizations and farmers’ groups established village-level nurseries for multiplying disease-free planting materials of improved varieties, mostly for farmers within their communities but for some in neighboring areas.

Discussions with researchers, implementation staff, and beneficiaries revealed that the program led to the following outputs:
• The formation of strong partnership and networking among researchers in IITA and INERA, FAO, SECID, CADIM, and PACT-Congo, farmers’ associations, farmers’ groups, and small- and medium-scale enterprises.
• Capacity building resulted in the build-up of knowledge, skills, and competencies at the individual and organizational levels for researchers in INERA, extension agents, farmers, farmers’ groups, processors, and equipment manufacturers. Farmers’ field schools helped farmers to gain experience.
• The development and release of disease-resistant improved varieties. When the program started there were no varieties resistant to EACMV-Ug but in 2–3 years, five varieties had been developed. Eleven additional varieties were released between 2005 and 2008.
• Crop management and crop protection technologies were delivered through breeding for disease and pest resistance, releasing predators for green mite control, and refining extension recommendations.
• Processing technologies and institutional innovations for organizing and linking farmers to markets were delivered after the adoption of improved crop management practices and expansion in cassava production in the targeted areas. Small and medium enterprises engaged in cassava processing emerged as a result of the improved processing technologies to expand their operations and market in micro-chips and other products, such as farinha, gari, and starch.

Second stage: During this stage the outputs were delivered to change agents, including INERA researchers, extension workers, NGOs, farmers’ associations, and private sector companies, resulting in changes in their level of awareness, knowledge, and practices.

The major outcomes perceived at the change agent level were as follows:
• Changes in the practice and behavior of INERA researchers, Government and NGO extension agents, processors, and equipment manufacturers.
• Development of supply systems for clean planting materials of improved varieties, advice on crop and postharvest technology management, locally manufactured processing machines and equipment, micro-enterprises engaged in cassava processing, improved quality of cassava flour, better output marketing and logistics of distributing cassava-derived products to urban consumers.
• Tissue culture and institutional arrangements for the multiplication and distribution of planting materials made possible the distribution of a cumulative total of 417,354,633 one-meter stem cuttings of disease-free improved varieties to 3,530,666 households from 2001 to 2008. However, the total planting materials distributed were sufficient for planting only a lower bound estimate of 166,942 ha, or about 9% of the total national cassava area cropped in 2007/2008 using FAO data or 10% using data from the Service National de Statistique Agricole (SNSA).

Third stage: Here, the research products were delivered to farmers, resulting in increased awareness, knowledge, and adoption of improved practices among farm households in villages exposed to the program’s interventions compared to those living in other villages. Respondents interviewed in this study believed that farmers exposed to the program’s interventions acquired new knowledge that mosaic is a disease. To get good yields, growers needed to obtain disease-free planting materials from INERA research stations or NGO multiplication plots instead of neighboring farmers. Farmers acquired knowledge on using phytosanitation to control the disease.

Fourth stage: Respondents perceived that R4D generated synergies between farmers’ access to and participation in markets, on-farm productivity, and the intensiveness in which parcels of land were cultivated. This, in turn, accelerated the adoption of improved technologies and farm-level impacts. Households in intervention villages that participated in the R4D program were perceived to have achieved better yields, higher profitability, and greater food security than those that did not. Processing added value to the cassava produced in targeted areas and the products were increasingly being sold to urban markets.

Bags of dried cassava chips in Mbuji Mayi, DRC. Photo by IITA.
Bags of dried cassava chips in Mbuji Mayi, DRC. Photo by IITA.

The study
Primary data were collected through a questionnaire interview survey in 2009 to a randomly selected sample of households in areas where the program had been implemented and neighboring nonprogram areas. The survey used stratified random sampling to select contact households. A total of 521 households clustered in 52 villages were interviewed.

The study tested three hypotheses on the impact of the agricultural R4D program on farm-level outcomes of interest:
• The R4D program has causal effects on households’ participation in markets.
• The R4D program generates synergies among improved varieties and crop management technologies and encourages their adoption by farm households.
• The R4D program helps households achieve higher plot-level yields, greater profitability, and improved household food security.

The study established that households in villages where R4D had been introduced had significantly higher levels of sales of cassava compared with households in villages without interventions. R4D was also found to increase the probability of a household adopting most of the technology options. There were high correlations among random utility components which provided evidence for the hypothesis that R4D generates synergies among improved technology adoption decisions.

The study showed that households who lived in intervention villages had significantly higher plot yields, gross margins, and food security than those in nonintervention villages. Marginal effects showed that household participation in a farmers’ organization had the most impact.

The study concluded that the cassava R4D interventions were successful in increasing the outcomes. The finding that the R4D program had positive impacts on intermediate outcomes that can be observed in the short term suggests that the approach has potential for long-term impacts on final outcomes. This implies that policymakers can increase the impact of agricultural research on household food security by promoting agricultural R4D approaches together with the development of farmers’ organizations.

Cassava is the second most important staple in Malawi after maize. The two crops supply over 70% of calories in the diet and sometimes replace and complement each other in production and consumption. They are historically intertwined as both were introduced into Southern Africa from Brazil in the 1500s. However, the colonial and early post-independence agricultural policies favored maize and, as a result, cassava production remained insignificant throughout this period.

Women peeling cassava for processing. Photo by IITA.
Women peeling cassava for processing. Photo by IITA.

The situation, however, changed dramatically between 1994–1995 and 2006–2007 when the area grown to cassava more than doubled and production expanded. This was as a result of a combination of factors including (1) realignment of commodity prices in favor of cassava over maize after the removal of consumer and producer subsidies under structural adjustment; (2) development, dissemination, and adoption of improved technologies; (3) extension to popularize cassava; (4) the collapse of input supply, credit, and maize markets; (4) a decline in soil fertility below the economic yields for maize; and (5) high rainfall variability.

The spread of HIV/AIDS may have also contributed by reducing the rural labor supply and replacing high-labor intensive crops such as maize with labor-saving, low-input crops such as cassava. There was an increased demand for fresh and processed cassava in central Malawi as consumers substituted cassava for more expensive maize and wheat products.

A study applied econometric modeling treatment effects methods to estimate the impact of the cassava R4D projects implemented in Malawi in the 1990s and 2000s on the farm-level yield, per capita area planted to cassava, and food security.

Overview of the program
Formal cassava improvement research started in 1930 in Karonga with the collection and evaluation of local varieties for their yield potential and resistance to cassava mosaic disease. Varieties from Malaysia, Java, Kenya, Tanzania, Trinidad, and Ghana were introduced and evaluated.

Notable highlights include the severe drought and famine in the years 1948–1949 and 1949–1950 that led the Department of Agriculture to distribute cuttings as a drought recovery intervention. However, these years were followed by three consecutive good rainfall seasons that resulted in sufficient maize production, large surpluses of cassava which were not sold due to lack of transport, and reduced interest in the crop except in areas where it was already a staple food.

During the 1950s and 1960s, researchers continued to search for mosaic-resistant varieties from the East African Agriculture and Forestry Research Organization. Agronomic trials were also conducted on intercropping, planting time and method, spacing, harvesting, fertilizer use and land preparation, and on pest and disease control. The findings were formulated into recommendations and made available to extension for dissemination to farmers.

Motorized cassava grating machine commissioned by IITA-CFC in Sierra Leone. Photo by IITA.
Motorized cassava grating machine. Photo by IITA.

Postharvest management research was conducted to develop technologies for processing cassava into flour on a large scale by agribusiness companies. In the 1950s and 1960s, the crop began to emerge as a cash crop in the southeastern districts when private traders exported surpluses of dried cassava to East Africa and the European Common Market. At its peak in 1968, the crop was the fifth highest foreign exchange earner in Malawi after tobacco, tea, groundnut, and maize. But the quality of the dried chips was low because of poor processing methods and could not compete with cassava pellets from Thailand.

From 1978, a parallel cassava research program was set up to evaluate materials from IITA. The breeding scheme used at IITA was introduced to shorten the time taken from identification to variety release, multiplication, and distribution of planting materials to farmers. This resulted in the release of second generation bitter varieties tolerant of cassava mosaic and mealybug in 2000. These were Mkondezi (MK91/478), Silira (TMS601428), and Maunjili (TMS91934). In 2002, the national cassava program released two other cassava mosaic- tolerant bitter varieties, Yizaso (CH92/112) and Sauti (CH92/077).

At the beginning of 1985 there was a serious outbreak of cassava mealybug in the main cassava-growing areas but scientists brought the pest under control in the 1990s by exporting and releasing its natural exotic enemies from IITA.

During 1991–1992 and 1993–1994 there were severe droughts followed by low rains in 1994–1995. In response, the national cassava research and extension programs expanded and accelerated the multiplication and distribution of planting materials for cassava and sweetpotato. This was followed by another food security project from 1998–1999 to 2000–2001 which also incorporated postharvest technologies.

The projects resulted in major changes in the organization and implementation of cassava research. The first change was the smart borrowing of IITA’s procedures for large-scale tissue culture, the rapid multiplication of virus-free planting materials, and distribution systems. The systems consisted of farmers’ groups, researchers, extension agents, traders, processors, religious groups, community based and nongovernmental organizations, and policymakers. Also involved were Bunda College of Agriculture, Natural Resources College, IITA/SARRNET, International Potato Center, FAO, United Nations Children’s Fund, and donors (Office of Foreign Disaster Assistance/United States Agency for International Development, United Nations Development Program, and International Development Research Centre.

The first multiplication and distribution project focused on the supply of ‘‘cleaned” cassava cuttings of improved varieties through a quality control system of primary, secondary, and tertiary nurseries, on-farm technology evaluation and dissemination of improved crop management practices, the development of farmers’ organizations, training, capacity building, and networking.

The second project placed more emphasis on postharvest management and market development. The components were implemented as a package in target areas selected as being food insecure suited to cassava production, and suitably located to minimize the costs of transporting materials from primary multiplication sites at government research stations, agricultural colleges, irrigation schemes, and agricultural training centers and from secondary sites in NGO intervention areas. The planting materials were supplied, based on availability and farmers’ requests, to villages through farmers’ groups and distributed through farmer-to-farmer exchange.

Research findings and conclusions
The study found that the cassava R4D program benefited smallholder farmers and generated significant farm-level impact. Using synthetic control methods to control for observable characteristics it showed that by 1995 annual yields in predominantly cassava-growing and -consuming districts first exposed to the program were about 23% higher than they would have been in the absence of the program.

The study estimated an increase of 14% in per capita area cropped to cassava among households first exposed to the program compared with those that were later exposed. The cassava R4D program led to an average increase of 9.1% for the 1997–1998 cross-section, 9.5% for the 2004–2005 cross-section, and 8% in the before and after changes for households per capita area planted to cassava.

Using the Heckman’s treatment effects model to control for observables and unobservables, the study estimated that participation in the program increased the months a household can meet its minimum caloric requirements from home-produced maize and cassava staples by 66% for a randomly selected household, 18% for those actually selected in the program, and 22% for those at the margin of participation.
Therefore, increasing the impact of cassava R4D at a greater scale requires further investments in an adequate supply of planting materials and market development to transform cassava into both a food and cash crop.

Outcome mapping: a tool for monitoring and evaluation

E.A. Ouma, and G.A. Neba,

IITAroutinely measures impact resulting from its R4D projects and programs. Photo by IITA.
IITA routinely measures impact resulting from its R4D projects and programs. Photo by IITA.

Many development practitioners are preoccupied with the identification and measurement of impact resulting from their research-for-development projects or programs. In many high-level meetings, the importance of results-based management that is goal-oriented and that emphasizes cause and effect of inputs, outputs, and impacts, has been emphasized and a large number of methodological guidelines have been developed.

One such guideline is the Logical Framework Approach (LFA). It is a hierarchical linear causal-effect chain presented at four levels (activities, outputs, outcomes, and impact). It is concrete and encourages the clear formulation of outcomes and goals/impact and the precise definition of quantifiable targets. Its major weakness is the attribution of cause and effect between the levels of outcome and impact (Jones 2006). In reality, this cannot be conclusively determined. Most impacts occur a long way downstream and may not be directly influenced by a single actor. In addition, the linear cause–effect thinking in LFA is a rather strong assumption and has been criticized by many practitioners.

The weaknesses in the existing tools, particularly in the monitoring and evaluation of developmental impacts, motivated the International Development Research Centre to develop a different approach, referred to as outcome mapping.

Figure 1. Boundary partner's link to the program and the real world.
Figure 1. Boundary partner's link to the program and the real world.

Outcome mapping
Outcome mapping is a method for planning and assessing the social effects and internal performance of projects, programs, and organizations (Earl et al. 2001). It helps a project or program team to be specific about its targets, the changes it expects to see, and the strategies it employs, and as a result, to be more effective in terms of the results it achieves. Results are measured in terms of changes in the behaviors of people, groups, and organizations, also known as “boundary partners” (Fig. 1) with which a project/program works directly. The project/program works with the boundary partners to effect a change but it does not control them.

The changes are referred to as outcomes. In so doing, outcome mapping clears away many of the myths about measuring impact and focuses more on social changes within complex and dynamic partnerships. Once boundary partners have been identified, outcome mapping differentiates the levels of behavioral change which may be seen among the partner organizations—known as progress markers. These are grouped according to expected behaviors (early positive responses), desired behaviors (active engagement), and hoped-for behaviors (deep transformation in behavior) (Shaxson and Clench 2011). In the vocabulary of outcome mapping, these are behaviors we would ‘expect to see”, “like to see”, and “love to see” and they may be priorities for change or a time sequence of activities, or a mixture of both (Fig. 2).

Figure 2. Progress markers of a boundary partner. Source: Jones 2006.
Figure 2. Progress markers of a boundary partner. Source: Jones 2006.

Attribution and measurement of downstream results are dealt with through a more direct focus on transformations in the actions of the main actors. The outcomes enhance the possibilities of developmental impacts but the relationship is not necessarily a direct one of cause and effect. The outcomes can be logically linked to a project’s activities although they are not necessarily directly caused by them. Outcome mapping, therefore, focuses on the contribution of a project in the achievement of outcomes rather than claiming the achievement of developmental impacts.

The development of M&E tools (both qualitative and quantitative) for assessing outcomes and impact on commodity systems, including outcome mapping and participatory impact pathway, was identified as an output target for IITA’s Opportunities and Threats Program in 2011 (IITA 2009). This highlights the importance of developing tools not only for documenting technology adoption trends and impact but also those that monitor outcomes, providing stakeholders with timely information about their progress and achievements for systematic and collective learning, reflection, and corrective action.

A few R4D projects at IITA have employed outcome mapping or some of its elements in their M&E framework. For instance, the Consortium for Improving Agriculture-based Livelihood in Central Africa project, largely operating in the East and Central African highlands, follows the spirit of outcome mapping in its arrangements to scale out technology. The boundary partners, comprising international and national NGOs and farmers’ associations, articulate their goals, expectations, and contributions through informal or formal memoranda of agreement with the project. The project endeavors to meet the partners’ expectations through jointly planned activities to achieve the expected outcomes, which have prospects of producing sustainable impacts.
Opportunities for interactions between a boundary partner and the project and among the boundary partners are made available for collective learning, to evaluate progress towards the achievement of goals over time, and to identify corrective measures.

Other CGIAR centers, particularly the International Center for Tropical Agriculture (CIAT), International Livestock Research Institute (ILRI), and the World Agroforestry Centre, apply outcome mapping in their natural resource management and livestock projects.

Stages of outcome mapping and monitoring tools
The process is divided into three stages. The first, referred to as the intentional design phase, is largely a planning stage. This helps a project to establish a consensus on the macro-level changes it will help to bring about and to plan the strategies it will use. It is based on the principle of participation and purposefully includes those implementing the project in the design and data collection so as to encourage ownership and use of the findings. It involves articulation of the vision and mission of the project, the identification of the boundary partners, the outcome challenges, progress markers, and strategies to be employed for changing the behavior of boundary partners to better deliver the progress markers. Supportive strategies facilitate change, possibly by one partner providing information, capacity, or skills to others.

The second stage is outcome and performance monitoring. It provides a framework for an ongoing monitoring of the projects’ actions and the boundary partners’ progress toward the achievement of outcomes. It is largely based on a systematized self-assessment and uses monitoring tools for elements identified in the design stage. The tools include an outcome journal (for monitoring progress markers), a strategy journal (for monitoring the strategy maps) and a performance journal (for monitoring the organizational practices).

The third stage is evaluation planning. It helps the project to identify evaluation priorities and develop an evaluation plan (this targets priority areas for detailed evaluation studies).

Strengths and weaknesses
Outcome mapping provides a focus on institutional transformation that is often lacking in techniques which emphasize the delivery of outputs as an indicator of achievement. It aligns itself with the realities of development, often occurring in complex and open systems with multiple actors. The methodology ensures the clear formulation of responsibilities, roles, and progress markers for each project partner in addition to providing a framework and the tools for continuous monitoring. Measurable outcomes and clear milestones enhance ownership by the local actors and beneficiaries as well as the management of multiple accountabilities (project, beneficiaries, partners, and donors).

Outcome mapping’s one-dimensional focus on “changes in behavior”, although important to sustainable development, cannot be an end in itself for development. The behavioral changes should support improvements in situations at a higher level. There is a need to have clear hypotheses about the framework, tools, and indicators for impact at the level of development results (such as the MDGs). Roduner et al. (2008) have proposed a synthesized model combining the strengths of outcome mapping focusing on capacity building and the logical framework with its focus on development results. The synthesized model has, however, not yet been tested.

Earl S, T Smutylo, and F Carden. 2001. Outcome mapping: Building learning and reflection into development programs. IDRC, Ottawa, Canada.

Jones H. 2006. Making outcome mapping work. Evolving Experiences from Around the World. IDRC, Ottawa, Canada.

Roduner D, W Schläppi, and W Egli. 2008. Logical Framework Approach and Outcome Mapping: A Constructive Attempt of Synthesis. A Discussion Paper, ETH, Zurich, Switzerland.

Shaxson L, and B Clench. 2011. Outcome mapping and social frameworks: tools for designing, delivering and monitoring policy via distributed partnerships. Delta Partnership working paper No 1,