ESTIMATING RICARDIAN MODELS WITH PANEL DATA

Although the Ricardian model is a cross sectional method, there are advantages to estimating the model with additional years of data. For instance, with a panel, one can more easily separate events in a single year (e.g. weather and price shocks) from longer term phenomenon such as climate. Many early studies used repeated cross sections to study panel data but one can get consistently better performance from panel methods. In this paper, we rely on two panel methods to estimate the Ricardian function for the United States across time. The results suggest that moderate warming scenarios would benefit American agriculture as a whole but more extreme climate scenarios would be damaging.