The Impact Of Temperature And Precipitation On County-level Economy In China Under A Changing Climate

The past few decades had witnessed significant changes in global climate,and the warming earth's surface and more frequent extreme weather events are especially notable.Hence,evaluating the impact of weather variations on economy and human society become a hot topic in economics and other fields.Lots of empirical studies unanimously find that weather shocks remarkably affects aggregate output,agriculture,industrial output,international trade,labor productivity,health,migration,conflicts,aggression and crime,energy consumption,etc.The climate in China also has experiencing notable changes since 1980 s.A noteworthy question is how the Chinese economy is affected by climate change.Employing county-level economics data and weather data from 1996 to 2012,we investigate the following three questions.Firstly,how extreme weather events influence county-level aggregate output and sector-level output;whether the non-linear relationship between weather variables and economic performance found by some empirical studies still holds in China.Secondly,we investigate whether changes in temperature and precipitation would influence a county's economic growth.Thirdly,we intend to explore the adaptation ability of Chinese economy to weather shocks and identify possible opportunities and challenges in the future.To answer these questions,we proceed the paper in the following way.Firstly,we define several basic concepts,including climate,weather,extreme weather(climate)events,climate change,and clarity the difference between climate and weather.Then we introduce weather data of different categories and their sources.Next we provide a theoretical model to describe how the impact of climate change can be correctly identified by weather variation.As preparation for the empirical studies,several popular climate econometric methods are also introduced.Secondly,we develop a two-sector model to show how weather factors influence agricultural and non-agricultural production.This model offers a way to add up daily weather factors to year-level and enables us to match the weather data to annual economics data.Then we examine the non-linear relationship between weather factors and county-level outputs.Our empirical results show that high temperatures(20-30?)and extremely high temperatures(>30?),compared with moderate temperatures(10-15?),have significantly detrimental effect on county-level economic outputs.Agricultural and non-agricultural sector responses to weather shocks in different ways.We find that agricultural production is more sensitive to extreme weather conditions,extremely high temperatures and extreme precipitation both reduce the grain outputs,cotton outputs,oiling crops outputs and agricultural income significantly.Generally,the China is more sensitive to extremely high temperatures relative to the United States.We also find regions which experiencing frequent extreme weather events have a better performance in adaptation and the crop production in agricultural counties is less sensitive to weather extremes.In addition,air-conditioning devices,to some extent,are capable of mitigating the adverse effect of extremely hot days.Finally,we examine the impact of weather shocks on a county's economic growth.In our theoretical framework,we allow weather variables affect both economic output and economic growth.In the empirical section,we adopt autoregressive distributed lag model to test the immediate effect and accumulative effect of weather shocks.Our results show that the impact of weather shocks on economic growth is permanent rather than temporary.Specifically,a one Degree Celsius increase in annual average temperature would generate an accumulative effect which decrease county-level economic growth rate by 3.82-6.04% in a decade;daily precipitation increasing by 0.1mm(i.e.,annual precipitation increasing by 36.5mm)would decrease accumulative county-level economic growth by 0.55-0.65%.Apart from agricultural sector,weather variations could also affect economic growth through other channels,such as labor productivity and investment etc.In addition,we also examine the relationship between weather changes and growth changes in the early and later periods in our dataset by the long difference approach,and we find that the impact of weather changes on economic growth over two separated periods is much larger than the impact of annual weather variations.Therefore,we conclude that the Chinese economy performs poorly in adaptation.Our study has three-fold contributions.First,we develop a highly flexible theoretical framework,which describes the direct and indirect channels of how weather changes influence economic production and growth.Second,we exploit a better dataset with high-frequent weather data and county-level statistic data.The high-frequent weather data enable us to establish series of temperature and precipitation bins and estimate the non-linear relationship between economic performance and weather factors.Moreover,the county-level economic data have more observations and more indicators,which ease the omitting variables biases and model misspecification,and help to identify the main channels that weather shocks affect economic output.Third,we apply various approach include constructing temperature and precipitation bins,dummy variables,interactions and exploiting the long different approach to estimate the non-linear effect of temperature and precipitation shocks on economic outcomes at county level.We also investigate the heterogeneity in responses among different regions and the overall adaptation ability of Chinese economy.