Characterization of maize production in Southern Africa: synthesis of CIMMYT/DTMA household level farming system surveys in Angola, Malawi, Mozambique, Zambia and Zimbabwe
In: http://hdl.handle.net/10883/1352
This report presents the synthesis of household level surveys in five intervention countries (Angola, Malawi, Mozambique, Zambia, and Zimbabwe) of the Drought Tolerant Maize for Africa (DTMA) project designed and implemented by the International Maize and Wheat Improvement Center (CIMMYT), International Institute for Tropical Agriculture (IITA) and national research and extension institutions in 13 countries of Sub-Saharan Africa (SSA). In each of the study countries, two districts were randomly selected provided that the districts fall in predetermined categories (20-40%) of probability of failed season (PFS). A total sample of 1108 households was randomly drawn with sample sizes varying country to country. The report has different sections that focus, in order, on description of the sample households, extent and determinants of poverty and inequality among the sample population, characteristics of maize production, perception and management of drought risk, and determinants of likelihood and intensity of adoption of improved maize varieties. The distribution of the age of the sample population shows that the population below the age of 16 years is 54% in Zambia, 47% in Malawi and more than 42% in the other three countries. Most of the sample households in each of the countries are headed by males. Only, Malawian sample has about one third of the households headed by women. The literacy level of household heads is considerably high by African standards. The proportion of literate household head ranges from 67% in Angola to 97% in Zimbabwe. The details of the literacy level show that about 48% of Zimbabweans have attended secondary school or higher followed by 32% in Zambia and to 28% in Angola. Most of the literate household heads in Malawi and Mozambique fall in the primary school category. The farming systems in the study areas are predominantly traditional and semi-subsistence oriented. The plough culture is an important feature of the systems and hence the traction power of draft animals is indispensable. Nonetheless, only Zambian (44%) and Zimbabwean (42%) farmers use draft animals. No sample household in Angola and Malawi owns a draft animal. The livestock owned per household, in tropical livestock units (TLU), ranges from 0.41 in Malawi to 2.9 in Zambia. Accordingly, the current value in US$ of the livestock owned by a typical household ranges from 102.7 in Malawi to 1051 in Zambia. The average farm land holding is highest in Mozambique where a typical household owns nearly 8 hectares of land, followed by Zambia at 6.63 hectares, and Zimbabwe at 3 hectares. The smallest average farmland was observed in Malawi with a typical household owning 1.25 ha. Wealth indices were computed based on asset holdings to look into the relative welfare distribution of the sample communities. Forty four percent of the sample households in Angola have negative wealth index. The households with negative indices can generally be considered as poor. Nearly 55% Malawian, 57% Mozambican, and 54% Zambian sample households do have negative wealth indices. Sixty two percent of the sample households in Zimbabwe have negative wealth indices, which is higher than any other country in the study.Asset wealth based rough classification of the households has shown that most of the sampling households are poor. A more detailed analysis of poverty and inequality was done based on reported income and expenditure. Household level determinants of poverty were identified using quantile regression. Generally, sample households in Mozambique and Zambia were found to be poorer than the sample households in Angola and Malawi. The poverty profiles show that the absolutely poor households in four of the countries (Angola, Malawi, Mozambique, and Zambia) do have significantly smaller family size, smaller number of illiterate household members, less number of important assets such as phones and radio, livestock and smaller farm sizes. An important observation is that the proportion of total land allocated to maize by absolutely poor households is significantly higher than that of better-off households. The study has also detailed the extent and determinants of poverty and inequality in the countries. The importance of maize technology use and resource allocation to the crop in determining magnitude of poverty and inequality is an important finding in view of the fact that the sample population is essentially semi-subsistent with limited market orientation. This finding also justifies the effort being exerted on development and deployment of maize and maize related technologies in rural communities of the study countries. Maize production in the region has peculiar characteristics with important distinctions across countries. The land allocated to maize ranges from 45.9% in Mozambique to 69.8% in Malawi of the whole farmland. In Angola, Malawi and Mozambique, most of the maize land is covered with land races; whereas in Zambia and Zimbabwe, hybrid maize covers most of the maize area. All the sample farmers in Zimbabwe are aware of the difference between improved Open pollinated varieties (OPVs) and hybrid maize varieties. On the contrary, about 95% of the sample farmers in Angola do not know the difference between OPV and hybrid maize. Most of the sample farmers in Malawi (72.0%), Mozambique (98.0%), and Zambia (78.0%) are in fact aware of the difference between OPVs and hybrid maize varieties. Regarding recycling of hybrid seeds, it was found out that Zimbabweans hardly recycle, whereas Mozambicans do on average recycle 1.5 times. This pattern of recycling also applies to improved OPVs. Despite considerable number of farmers depending on the market to fetch maize seeds, 90.7% of the farmers in Angola purchased and planted only local maize varieties. Malawian and Zambian farmers, followed by Mozambicans and Zimbabweans, do mostly purchase and grow improved seeds. In terms of the proportion of seed types used, Zimbabwe stands out well-above others with 94.6% of the seed used being improved, followed by Zambia (64.3%) and Malawi (24.8%). In Malawi, the most preferred varieties, in order of preference are: local, MH36, Kanyani, and Makolo. In Mozambique, Ndau ou Chindau, Matuba, SC513, Laposta, and Pan 67 were indicated to be the most preferred varieties in that order. In Zambia, the four most preferred varieties were identified to be Gankata, SC513, Pool 16, and Obatampa. In Zimbabwe, SC513 is the most preferred variety followed by ZM521 and the local Heckory King variety. Drought was reported to be the most important challenge on the livelihoods of people in Malawi, Zambia and Zimbabwe, whereas it was indicated to be second, next to sickness and mortality of a family member, in Angola and Mozambique. Maize varieties in general and improved OPVs and hybrids in particular are being considered very risky in terms of predictability and reliability of yields. Given the importance of maize and the vulnerability of the farming communities in the region, drought and risks associated with it will have paramount and potentially irreversible consequences on the poor sections of the region. The decisions regarding level and intensity of improved maize adoption have also been investigated to show that gender based intra-household division of labor was an important factor considered in deciding to adopt or not improved maize varieties. Asset endowments such as farm size and livestock wealth were found to be important determinants of level and intensity of adoption in the region. Similarly, membership in social groupings and engagement of off-farm activities influenced adoption decisions. Access to extension services has universally been identified as an important factor in determining the level and, when relevant, the intensity of adoption of improved maize varieties in the region. As important as this service is, however, the extent to which farmers are getting the service is not that encouraging according to respondents. It is, therefore, imperative to underline again the need for investment in the agricultural extension system and the effort that shall be exerted in enabling the private sector to engage in generation and deployment of agricultural information. In designing and implementing any intervention that aims at contributing to the risk coping ability of farmers, it is essential to take into account heterogeneity of the farming communities. Due consideration of this heterogeneity shall be made while assessing the importance of drought risk and while analyzing the effectiveness of the contributions to be made with the intention of strengthening drought risk coping strategies. The trait preferences of farmers are for instance an important indicator of the heterogenous demand structure. This study has shown that despite the fact that yield size is among the most preferred traits, farmers' strong reference to maize as a risky crop urges refocusing breeding activities to generation of germplasms with reliable yield distribution. Farmers have also shown strong interest in traits such as drought tolerance, early maturity, and good performance under poor rainfall implying the need for multi-trait focused breeding schemes. Public agricultural extension institutions and public mass media are by far the two most important sources of agricultural information in the region. Despite the political importance of agriculture in general and maize in particular, there is always a lack of incentive in publicly owned institutions to deliver the information as timely and as adequately it is needed. Agricultural extension efforts in the region should in fact be accompanied by comprehensive microfinance institutions to relieve farmers of the seasonal cash shortage which almost all of them experience every year. So far, except in Zimbabwe, access to rural credit and finance seems to be farfetched.