This article explores the economics of input decision under production uncertainty. The article develops a methodology to specify and estimate cost-minimizing input decisions under a state-contingent technology. The analysis is applied to time series data on U.S. agriculture. It finds strong empirical evidence that expected output alone does not provide an appropriate representation of production uncertainty. The results provide empirical support for an output-cubical technology. This indicates that an "ex post" analysis of stochastic technology appears appropriate. The analysis also provides evidence that the cost of facing production risk has declined in U.S. agriculture over the last few decades. Copyright ©2007 American Agricultural Economics Association.