Identification of Counterfactuals and Payoffs in Dynamic Discrete Choice with an Application to Land Use

摘要: Dynamic discrete choice models are non-parametrically not identified without restrictions on payoff functions, yet counterfactuals may be even when payoffs not. We provide necessary and sufficient conditions for the identification of a wide range with nonparametric payoffs, as well commonly used parametric we obtain both positive negative results. show that access to extra data asset resale prices (when applicable) can solve non-identifiability counterfactuals. The theoretical findings illustrated empirically in context agricultural land use. First, results unobserved market-level state variables. Then, using unique spatial dataset use choices prices, estimate model investigate two policy counterfactuals: long run elasticity (identified) fertilizer tax (not identified, affected dramatically by restrictions).