This paper explores the economics of input decision by a firm facing production uncertainty. It relies on a state-contingent approach to production uncertainty. First, the paper develops a methodology to specify and estimate cost-minimizing input decisions under a state-contingent technology. Second, the analysis is applied to time series data on US agriculture. It finds strong empirical evidence that, in the analysis of input choices, 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 (as commonly found in previous research) appears appropriate. The analysis also provides evidence that the cost of facing adverse weather conditions has declined in US agriculture over the last few decades.