Second, we comprehensively analysed health, environmental, social and economic impacts and related trade-offs and synergies for a number of future scenarios of Swiss agricultural production and food consumption. For this, we used a modelling approach, linking three different models: a global mass flow model, a system dynamics model and an environmentally extended input-output model. We modelled ten different scenarios for the Swiss Food Sector in 2050. These scenarios were either developed in a participatory process during a series of interviews and group discussions with different groups of stakeholders or optimised environmental impacts while at the same time complying with different nutritional and agronomic restrictions. Three main scenarios were analysed with all three models in detail. Among these main scenarios was the SwissFoodPyramid2050 Scenario, which assumes a widespread implementation of the nutritional recommendations according to the Swiss Food Pyramid. The FeedNoFood2050 Scenario assumes an improved use of agricultural land by feeding only grass and by-products to livestock, which was not competing with direct human nutrition, i.e. did not require arable land (neither in Switzerland nor abroad). The third scenario was a reference scenario, which assumes no changes in diets until 2050 and which was used to compare the two alternative scenarios. The other scenarios were targeted at specific questions such as minimizing greenhouse gases. Our results illustrate two visions of how healthy diets and sustainable food systems could look like. Both the SwissFoodPyramid2050 and the FeedNoFood2005 scenarios would require similar dietary changes, such as a reduction of meat consumption and an increase of consumption of pulses. However, there are also fundamental differences between the diets in the two alternative scenarios, e.g. regarding the type of meat consumed. These differences can be interpreted as trade-offs which result from agronomic boundary conditions such as the coupled production of milk and meat, the availability of natural resources, such as grassland and co-products of food processing and health aspects of Swiss diets. Of primary importance in this respect was the use of permanent grasslands and the co-production of veal and beef with dairy production due to environmental reasons and reasons for optimally utilizing available resources. This means, if permanent grassland should be maintained as an ecosystem, dairy production would provide the basis for animal proteins. Thus, while in the FeedNoFood2050 Scenario veal and rather low-quality beef from dairy cows is consumed instead of meat from monogastrics, the SwissFoodPyramid2050 Scenario would result in a higher amount of meat from monogastrics. Our results imply that there is a lack of a comprehensive food systems view in the current discussion on healthy and sustainable diets. Stronger coherence between health, food and agricultural policy is needed to account for systemic boundary conditions and thus to allow for minimising trade-offs and maximise synergies. Current agricultural policies fail to address the health perspective. Financial support for meat and sugar producers, which lead to lower prices for those products and ultimately to a higher consumption than without these policies, are two obvious examples. Yet, comprehensive visions such as the SwissFoodPyramid scenario, the FeedNoFood Scenario or optimised scenarios would require an even more complex policy mix of incentives, regulations and information campaigns. This would probably need an adaptation of the current institutional setting and division of competences between the Federal Offices for Agriculture (FOAG) and for the Environment (FOEN), the State Secretariat for Economic Affairs (SECO) and the Federal Food Safety and Veterinary Office (FSVO). A commonly shared vision, including specific goals with respect to how the Swiss food system should look like, is urgently needed. Developing such a vision needs to involve all operators and stakeholders of the food system, as our results imply that more sustainable and healthy diets do not necessarily go along with financial benefits of both producers and consumers. These trade-offs and the knowledge of behavioural economics need to be considered for designing settings which create mutual benefits for operators in the food sector. For instance, neither the majority of consumers, food industry nor agricultural producers can be expected to respond altruistically as an entire sector in the long term. Therefore, policy needs to set financial incentives for internalising environmental and social externalities in order to push and pull the food system towards sustainability. Furthermore, it is crucial to account for agronomic boundary conditions and systemic aspects, such as the role of ruminants in utilizing grasslands and the unavoidable link of milk and meat production. Se hele rapporten under "relevante dokumenter".