Working Paper: NBER ID: w13799
Authors: Wolfram Schlenker; Michael J. Roberts
Abstract: The United States produces 41% of the world's corn and 38% of the world's soybeans, so any impact on US crop yields will have implications for world food supply. We pair a panel of county-level crop yields in the US with a fine-scale weather data set that incorporates the whole distribution of temperatures between the minimum and maximum within each day and across all days in the growing season. Yields increase in temperature until about 29C for corn, 30C for soybeans, and 32C for cotton, but temperatures above these thresholds become very harmful. The slope of the decline above the optimum is significantly steeper than the incline below it. The same nonlinear and asymmetric relationship is found whether we consider time series or cross-sectional variation in weather and yields. This suggests limited potential for adaptation within crop species because the latter includes farmers' adaptations to warmer climates and the former does not. Area-weighted average yields given current growing regions are predicted to decrease by 31-43% under the slowest warming scenario and 67-79% under the most rapid warming scenario by the end of the century.
Keywords: Climate Change; Crop Yields; Nonlinear Effects; Temperature
JEL Codes: C23; Q54
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| temperature (Y60) | crop yields (Q15) |
| temperature (beyond critical thresholds) (Q54) | crop yields (Q15) |
| temperature > 29°C (Y50) | corn yields (Q11) |
| temperature > 30°C (Q54) | soybean yields (Q16) |
| temperature > 32°C (Y50) | cotton yields (Q15) |
| extreme temperature events (Q54) | crop yields (Q15) |