This paper evaluates the effect of agronomic uncertainty on bioenergy crop production as well as endogenous commodity and biomass prices on the feedstock composition of cellulosic biofuels under a binding mandate in the United States. The county-level simulation model focuses on both field crops (corn, soybean, and wheat) and biomass feedstocks (corn stover, wheat straw, switchgrass, and miscanthus). In addition, pasture serves as a potential area for bioenergy crop production. The economic model is calibrated to 2022 in terms of yield, crop demand, and baseline prices and allocates land optimally among the alternative crops given the binding cellulosic biofuel mandate. The simulation scenarios differ in terms of bioenergy crop type (switchgrass and miscanthus) and yield, biomass production inputs, and pasture availability. The cellulosic biofuel mandates range from 15 to 60 billion liters. The results indicate that the 15 and 30 billion liter mandates in the high production input scenarios for switchgrass and miscanthus are covered entirely by agricultural residues. With the exception of the low production input for miscanthus scenario, the share of agricultural residues is always over 50% for all other scenarios including the 60 billion liter mandate. The largest proportion of agricultural land dedicated to either switchgrass or miscanthus is found in the Southern Plains and the Southeast. Almost no bioenergy crops are grown in the Midwest across all scenarios. Changes in the prices for the three commodities are negligible for cellulosic ethanol mandates because most of the mandate is met with agricultural residues. The lessons learned are that (1) the share of agricultural residue in the feedstock mix is higher than previously estimated and (2) for a given mandate, the feedstock composition is relatively stable with the exception of one scenario.