Despite the existence of a large and growing literature on the potential impact of climatechange on agriculture, there still exists some disagreement about the magnitude and even thesign. Our own research suggests that the impact on U.S. agriculture is likely to be stronglynegative, based on a series of studies in which we link farmland values to climate variables,and crop yields to both climate and yearly weather variables. Results are significant, robust,and consistent across data sets and methods. A recent but influential study by Deschˆenes andGreenstone (2007b) reports dramatically different results: based on regressions of agriculturalprofits and yields on weather variables, they conclude that the impact of climate change willbe either insignificant or positive. In this paper we reconcile these conflicting results.Likely explanations for the divergence between our findings and theirs are: (1) missingand almost certainly incorrect weather and climate data in their study, amplified by theuse of state-by-year fixed effects that absorb most year-to-year weather variation but leavedata errors intact; (2) their unusual and in our judgment incorrect treatment of climatechangepredictions; (3) their use of the older Hadley II climate model for climate changepredictions rather than the more recent and less optimistic Hadley III climate model usedin the Fourth IPCC Report and in our studies; and (4) theoretical difficulties in their profitbasedapproach due to the confounding effects of storage and possibly also capital andinventory adjustments and local price movements associated with weather fluctuations. Acareful account of these factors shows that the balance of evidence weighs heavily on theside of severe adverse potential impacts on U.S. agriculture by the end of the century, andprobably sooner, stemming from anticipated global warming.
