The allocation of variable inputs among crops is a common problem in applied studies that use farm accountancy data. Standard farm accounting information is typically restricted to aggregate or whole-farm input expenditures; there are usually no details on how these expenditures are split among crops. Most studies employing multi-crop econometric models with land as an allocable fixed input consider generally variable input uses at the farm level (Moore and Negri, 1992). However, the allocation of variable inputs among crops appears to be useful for several objectives, such as to analyze the evolution of gross margins at the crop level, to investigate the empirical validity of a multi-crop econometric model and to provide important information for extension agents or farmer advisors.