| Abstract: | A beef simulation model was used to examine a range of production possibilities for increasing the output of extensive ranching systems in the tropical savannas of eastern Colombia through the use of sown grass-legume pastures of higher inherent digestibility than that of the native species. Generalised stochastic dominance was used to compare some promising alternatives with the unimproved savanna-based production system and with each other. As the planning horizon of the decision maker was reduced, so variability, measured as the coefficient of variation of output, increased significantly. The impact of risk on the long-term acceptability of a complete investment was limited. |
