PhD. Thesis 2018Sustainability of smallholder dairy and traditional cattle milk production systems in developing countries, including Tanzania, is limited by a number of constraints such as low cow productivity, shortage of feed, limited access to inputs and outputs markets and degradation of natural resources. Efforts have been made to improve the sustainability, but the improvement is hindered by lack of knowledge on how to ensure sustainability of the production systems particularly at the farm level. To contribute to the efforts being made to address these issues, this study aimed at assessing sustainability of smallholder dairy and traditional cattle milk production systems in Tanzania. The study was conducted in four districts located in Morogoro and Tanga Regions. In the context of this study, a smallholder dairy production system refers to a system with dairy farms which have up to five dairy cows, where majority are crossbreeds of local and pure exotic breeds and milk is considered the main source of income. Meanwhile, a traditional cattle milk production system consists of cattle farms keeping mainly indigenous cattle and milk is not considered the main source of income. The first step of the study involved identifying relevant indicators for assessing sustainability of smallholder dairy as well as traditional cattle milk producing farms. The systems were further categorised into Rural production to Rural consumption (R-to-R) and Rural production to Urban consumption (R-to-U) systems. Whereby R-to-R refer to rural farmers sold milk to rural consumers and R-to-U to rural producers predominantly selling milk to urban consumers. A two-round Delphi approach involving 44 diverse experts and stakeholders was used in identifying the sustainability indicators. The second step involved developing a milk production farm sustainability assessment tool based on a set of fifteen most relevant of the identified indicators. The indicators were selected from the previously identified indicators according to data availability and cost. The tool was used to assess sustainability of 431 randomly selected farms in the study districts. The data were collected through interview of the farmers using a pre-tested questionnaire administered to the selected farms. Individual indicators of sustainability were measured, normalized using mini-max approach, weighted using factor analysis and aggregated into economic, social, environmental and overall sustainability indices using linear aggregation. The sustainability performance indicator and index scores were ranked from 0 to 1 and grouped into three categories of sustainability indicator / index scores namely The R-to-R systems and the R-to-U systems using a two-tailed Student’s t-test. The third step involved analysis of the relationships between the farm and milk producers’ organisations (POs) sustainability performances. The differences between farm sustainability mean performance indicators and indices for PO-member farmers and non-PO-member farmers were analysed using a two-tailed Student’s t-test. The sustainability of POs was assessed using an existing tool, “Producers’ Organisation Sustainability Assessment tool (POSA)”, which is based on a set of six economic and organizational dimensions. The relationships between farm and PO sustainability performance indicators were established using Pearson correlation analysis. The correlation coefficients (r) were categorized as weak (r < 0.3), moderate (0.3 ≤ r < 0.5) and strong (r ≥ 0. 5). Lastly, the study analysed the determinants of smallholder dairy and traditional cattle milk production farm sustainability. Descriptive statistics were analysed to understand the socio-economic characteristics of milk production farms. Then the socio-economic characteristics were compared between R-to-R systems and R-to-U systems using two-tailed Student’s t-test and chi-square for the means and proportions respectively. The double censored Tobit regression model was applied to analyse the determinants of farm sustainability. The Delphi technique refined an initial set of 57 indicators to a final set of 29 relevant indicators. The relevant indicators included 18 economic, seven environmental and four social indicators. Specifically, the key economic indicators were milk hygiene, cow productivity, income per litre of milk and access to milk market. Social indicators included participation in organizations, women’s empowerment and the education level of the farm manager; while environmental indicators were water conservation and access to water. Results from the farm sustainability assessment show that the economic mean score (0.27 ± 0.20), social mean score (0.32 ± 0.27), environmental mean score (0.31 ± 0.22) and overall mean score (0.30 ± 0.15) of farm sustainability indices were weak. The economic, social and overall sustainability mean performance index scores were significantly higher in the R-to-U systems than in the R-to-R systems (p < 0.05), implying better sustainability of R-to-U systems than R-to-R systems. The overall farm sustainability mean performance index, and its economic and social dimensions scores were significantly higher (p < 0.05) in PO-member farmers than in non-PO-member farmers. The “access to dairy production inputs and services” dimension of POs presented strong positive correlations with the overall farm sustainability performance index and its economic dimension (r = 0.58 and 0.67 respectively; p < 0.01). Similarly, the “access to dairy production inputs and services” of POs showed strong correlations (r = 0.70; p < 0.01) with cow productivity performance indicator. The farmers in R-to-U systems had significantly (p < 0.05) smaller land and herd size than in R-to-R system. Stall feeding system was the determinant factor (β = 0.256; p < 0.01) of economic sustainability. The determinant factors for social sustainability were stall feeding system (β = 0.165; p < 0.01), age of household head (β = 0.003; p < 0.05) and acquiring credit (β = 0.190; p < 0.01). The factor influencing environmental sustainability was stall feeding system (β = 0.098; p < 0.01). The factors influencing the overall sustainability were stall feeding system (β = 0.161; p < 0.01), the age of the household head (β = 0.001; p < 0.01) and acquiring credit (β = 0.081; p < 0.01). From the results of the study, it is concluded that a large number of existing indicators like greenhouse gas emissions could be considered less relevant in the context of Tanzania’s smallholder dairy and traditional cattle system than in other contexts. The study showed that 29 out of 57 sustainability indicators assessed were relevant to the studied system. The indicators identified here demonstrate the importance of matching any set of indicators to the characteristics of the specific production system being examined. The study provided a tool and framework for assessing sustainability of milk production farms in smallholder dairy and traditional cattle milk production systems in Tanzania using a set of 15 most relevant sustainability indicators out of the selected 29 indicators. The most relevant economic indicators were milk hygiene and cow productivity; social indicators were participation in organizations and women’s empowerment; environmental indicators were access to water and water conservation. Regarding the level of sustainability of the milk production farms, the results showed that the sustainability performances of smallholder dairy and traditional cattle milk production farms in the selected districts were weak, particularly in R-to-R system. Producers’ organisation sustainability performances, particularly its provision of dairy inputs, have strong positive relationship with farm sustainability performances, particularly the farm economic dimension. Indeed, stall feeding and access to credit tend to improve farm sustainability. From the results of the study, continued private and public investments in the non- traditional dairy areas and promotion of market linkages to urban areas where milk demand is stronger, is recommended not only for immediate improvement of livelihoods but also for sustainability considerations. Indeed, intensive dairy systems should be encouraged for higher sustainability of milk production and this could be possible by improving access to inputs and embedded services. The developed framework can be used by farmers, policy and decision makers to enable them identify key strengths and weaknesses and make respective decision towards sustainable milk production during implementation of dairy improvement programs.