For the vast majority of professional economists, it is quite natural to think of the elasticity of supply with respect to price as being positive. The same idea is a matter of controversis among agricultural economists. Often, it is alleged that the rationality is radically different for peasants and capitalists entrepreneurs. It is nevertheless possible to give account of observed facts within the strict neoclassical framework, but this requires a long run, dynamic point of view, and the explicit intervention of financial and security constraints at farm level. Various consequences for the estimation of supply functions may be derived. Especially, because the elasticity of supply is likely to vary greatly across time, it is difficult, and probably misleading to make use of a large number of observations from the past in order to estimate them by regression. Therefore, the recourse to statistical inference should be kept to a minimum level when estimating responses or aggregate production functions.
