This paper examines ex-ante impacts of two policy interventions that improve productivity of local-breed cows through artificial insemination (AI) and producers’ access to distant markets through a dairy market hub. The majority of cattle in Kilosa district in Tanzania are local low productivity breeds kept by smallholders and agro-pastoralists. Milk production is seasonal, which constrains producers’ access to distant urban markets, constrains producers’ incomes and restricts profitability in dairy processing. We developed and evaluated an integrated system dynamics (SD) simulation model that captures many relevant feedbacks between the biological dynamics of dairy cattle production, the economics of milk market access, and the impacts of rainfall as an environmental factor. Our analysis indicated that in the short (1 year) and medium (5-year) term, policy interventions have a negative effect on producers’ income due to high AI costs. However, in the long term (5+ years), producers’ income from dairy cattle activities markedly increases (by, on average, 7% per year). The results show the potential for upgrading the smallholder dairy value chain in Kilosa, but achievement of this result may require financial support to producers in the initial stages (first 5 years) of the interventions, particularly to offset AI costs, as well as additional consideration of post-farm value chain costs. Furthermore, institutional aspects of dairy market hub have substantial effects on trade-offs amongst performance measures (e.g. higher profit vs. milk consumption at producer's household) with gain in cumulative profit coming at the expense of a proportional and substantial reduction in home milk consumption.

Agriculture must face a number of very pressing environmental issues. We used the prototyping method to design three innovative cropping systems, each satisfying three ambitious goals simultaneously: (1) overcoming a major environmental constraint, which represents a major break regarding objectives to be reached in current cropping systems (differing between systems: a ban on all pesticides but with chemical nitrogen (N) fertilizer permitted; reducing fossil energy consumption by 50%; or decreasing greenhouse gas (GHG) emissions by 50%), (2) meeting a wide range of environmental criteria and (3) maximizing yields, given the major constraint and environmental targets. A fourth cropping system was designed, in which the environmental and yield targets were achieved with no major constraint (the productive high environmental performance cropping system (PHEP) system). The performances of these innovative cropping systems were compared to a conventional system in the Ile-de-France region. We used a three-step prototyping method: (1) new cropping systems were designed on the basis of scientific and expert knowledge, (2) these system prototypes were assessed with tools and a model (ex ante assessment) adjusted to the set of constraints and targets, with optimization by an iterative process until the criteria were satisfied and (3) evaluation in a long-term field experiment (ex post assessment), which is currently underway. We describe only the first two steps here, together with the results of the prototypes assessment with tools and a model. The pesticide, energy and GHG constraints were fulfilled. All these innovative systems satisfied environmental criteria in terms of nitrogen and phosphorus management, pesticide use, energy consumption and crop diversity. For the pesticide-free system, the soil organic matter indicator was lower than expected due to frequent plowing (every 2 years) and yields were 20–50% lower than for the PHEP system, depending on the crop considered. We focus our discussions on the design methodology and the availability of scientific knowledge and tools for projects of this type.