基于理性预期理论的大气污染经济成本评估  

Economic cost of air pollution:a rational expectation approach

作　　者：梅应丹 高立 邱筠 陈家德 MEI Yingdan;GAO Li;QIU Yun;CHEN Jiade(School of Economics and Management,China University of Petroleum(Beijing),Beijing 102249,China;Institute for Economic and Social Research,Jinan University,Guangzhou Guangdong 510632,China)

机构地区：[1]中国石油大学(北京)经济管理学院,北京102249 [2]暨南大学经济与社会研究院,广东广州510632

出　　处：《中国人口·资源与环境》2021年第2期24-33,共10页China Population,Resources and Environment

基　　金：国家自然科学基金青年项目“大气污染经济损失评估及有效治理的政策选择”(批准号:7170030429);“基于田野实验的中国农村秸秆还田‘助推式’干预策略研究”(批准号:71803192);广州区域低碳经济研究基地资助项目。

摘　　要：我国大气污染形势严峻,科学合理地评估大气污染的经济损失不仅有益于政策效益分析,同时是“绿色国民经济”核算的一项基础性工作。基于我国2013年2月到2018年7月的区县月度房屋交易数据和7种大气污染指标(AQI、SO_(2)、NO_(2)、CO、O_(3)、PM_(10)、PM_(2.5))的浓度数据,运用特征价格模型实证量化大气污染物减少的边际支付意愿(MWTP)和总经济损失。首先通过改变理性预期的时间段验证理性预期假设的成立,在此基础上采用理性预期方法解决遗漏变量所导致的内生性问题。研究结果显示:①NO_(2)、CO、PM_(2.5)和PM_(10)每上升1μg/m^(3),房价分别降低约2.04%、0.028%、0.34%和0.39%;而SO_(2)与O_(3)对房价的影响并不显著。②近年来大气污染的经济损失有所降低,政府的治理效果显著,但仍不容小觑。2013年AQI、PM_(10)和PM_(2.5)未达标导致的经济损失分别约为35600亿元、19300亿元和24100亿元,约占当年GDP的6.06%、3.29%和4.11%;2018年分别降低至19200亿元、5300亿元和6700亿元,占当年GDP的2.14%、0.60%和0.74%。尽管PM_(10)和PM_(2.5)浓度也在逐年下降,但仍未达到《环境空气质量标准》所要求的二级限值。最终评估结果显示,PM_(10)和PM_(2.5)二者导致的经济损失的加总数值,与AQI得到的数值相差无异。进一步证实了我国当前大气污染导致的社会经济福利损失主要是来自PM_(10)和PM_(2.5)的超标,因此治理“雾霾”是改善当前空气质量的关键。China is now confronted with severe air pollution issues.Scientific and reasonable estimation of the economic loss of air pollution is not only conducive to cost-benefit analysis of air pollution related policies,but also serves as a fundamental task for green national economic accounting.This study adopted the hedonic price model to estimate the non-market value of air pollution in China with the monthly county-level housing price index ranging from February 2013 to July 2018 and emission concentration data of seven major air pollutants(AQI,SO_(2),NO_(2),CO,O_(3),PM_(10),PM_(2.5)).For the identification strategy,the validity of rational expectation assumption was first tested and then a rational expectation approach was implemented to control for omitted variable bias.It assumed that the omitted housing attributes evolved based on a first-order Markov process.The estimated results suggest that:①A one unit(1μg/m^(3))increase of the concentration level of NO_(2),CO,PM_(2.5) and PM_(10) significantly lowered housing prices by about 2.04%,0.028%,0.34%and 0.39%,respectively.②The economic loss of AQI,PM_(10) and PM_(2.5) failing to meet the air quality standards was RMB¥3562 billion,RMB¥1933 billion and RMB￥2414 billion in 2013,accounting for 6.06%,3.29%and 4.11%of the GDP in China of 2013.It dropped to RMB¥1918 billion,RMB¥533 billion and RMB¥6676 billion in 2018,accounting for 2.14%,0.60%and 0.74%of the GDP in China of 2018.Although the concentrations of PM_(10) and PM_(2.5) were decreasing gradually,they still failed to meet the secondary limit value required by the Ambient Air Quality Standard.The estimated results indicate that the sum of the economic losses caused by PM_(10) and PM_(2.5) was roughly equal to that of AQI.This further confirms that the current social and economic welfare loss caused by air pollution in China mainly comes from the excess emissions of PM_(10) and PM_(2.5),so the key step to improve air quality is to reduce haze.

关 键 词：大气污染 理性预期方法 特征价格模型 经济损失 

分 类 号：F272[经济管理—企业管理]