Files
Abstract
A new procedure is developed to estimate innovation possibility frontiers and test for biases in technological change. Using data on four inputs (land, machinery, chemicals and labour) from central Canada (Ontario and Quebec) over the period 1926-1985, we find that the innovations possibilities frontier shifts neutrally over time. This is consistent with Ahmad's model of induced innovations, but is not consistent with de Janvry's application of Ahmad's model to the historical development of Argentine agriculture. Agricultural research in Canada has been conducted with the objective of developing cost minimizing technologies. Empirical support was found for this notion in the development of the innovation possibilities frontier.