Dynamic Panel Data Analysis of the Impacts of Climate Change on Agricultural Production in Japan

Abstract

This study empirically identifies the impact of global-warming-induced climate change on Japan’s agricultural production using panel data. First, we constructed panel data; combining time-series data from 1995 to 2006 for a cross-section of eight regions in Japan. Next, we conducted a static panel data analysis, using a function for agricultural products incorporating labor and three weather variables (temperature, solar radiation, and precipitation.) From the estimation results of the production function, we selected the production function with the aforementioned labor and weather variables and found that the rising temperatures and precipitation and falling solar radiation caused by climate change have reduced the rice production, while rising temperatures and precipitation have reduced the vegetable and potato production in Japan. Second, we conducted dynamic panel data analysis, using a production function for agricultural products incorporating labor, a one-period lagged output, and the same three weather variables. Based on the estimation results of the dynamic panel data model, we selected the production function for agricultural products using only the labor and three weather variables and found the same results for both the rice production and vegetable and potato production in Japan. Based on the estimated results of the static and dynamic-panel data models for variable mean annual temperature, which serves as a proxy for climate change, we concluded that an increase of 1°C in mean annual temperature would reduce rice production by 5.8% in the short term and 3.9% in the long term, and vegetables and potatoes productions by 5.0% and 8.6% in the short term and long term, respectively.