Geographic aggregation and induced innovation in American agriculture

Abstract

The induced innovation hypothesis is tested for the USA and western regions using cointegration procedures. An error correction model separates short-run and long-run effects of relative price changes. A significant difference in the elasticities of factor substitution along the isoquant and the innovation possibility curve implies induced innovation. The estimated results support the hypothesis for Washington, the Pacific Northwest, and the Western Region, but not for the nation. Corroborative tests of weak exogeneity fail to support the hypothesis in any of the geographic units. Changes in output level and research investment do not significantly bias agricultural technology in the USA.