Files
Abstract
In this paper, we propose a framework based on micro-level dynamic land use
models to predict the adoption of cover crops in the Upper-Mississippi River Basin. We
use preferences recovered using a dynamic discrete model of crop choice to build a dynamic
optimization framework to evaluate a range of scenarios based on the cover crops’ effect
on cash crop yields, costs of cover crop operations, and government support. We use a
conditional choice probability method to estimate the dynamic crop choice model based on
field-scale panel data and value function iteration method to assess counterfactual cover
crop scenarios. The framework is expected to be applicable to modeling decisions to adopt
conservation technologies in the absence of individual-level adoption data or for cases when
conservation technology is new. The dynamic crop choice model yields expected results
and reveals preferences for net revenues in line with previous literature. Simulation results
predict baseline cover crop adoption rates which, although in line with some recent farmer
surveys, are quite a bit higher than rates reported in 2012 Census of Agriculture. We
attribute these results to substitution patterns implied by a dynamic logit model estimated,
and suggest using aggregated Census of Agriculture data on state-wide adoption of cover
crops to calibrate the constants in the the estimated dynamic logit model as a possible
remedy under paucity of data related to individual decisions on cover crops adoption.