Spatial Analysis of Environmental Regulations: Evidence from Chinese Provinces

This research focuses on the spatial distribution and spatial correlation of environmental regulations (ERS) with exploratory spatial data analysis technique (ESDA). The panel data of 31 provinces in China during the period of 2005-2015 is used. Finds show that the most intensive ERS region is located around Beijing-Tianjin-Hebei region. The values of global Moran’s I vary from 0.1104 to 0.2145, indicating that ERS in China is not distributed randomly, but has spatial agglomeration effects. The High–High agglomeration type is mainly distributed in north China.


Introduction
Recently, a satellite photo, taken by NASA in 2013 releasing the latest distribution of China's haze-fog pollution, shocked all Chinese citizens. Rapid industrialization, urbanization and economic growth have resulted in an increase in energy consumption, air pollution, and associated health effects [1]. The China National Environmental Monitoring Center (CNEMC) reported that, in 2015, about 265 out of 338 cites could not meet the National Ambient Air Quality Standards of China, accounting for 78.4%. The population-weighted mean of PM2.5 in Chinese cities reached 61 μg/m3, about 3 times as high as global mean [2]. These showed the fact that not only China's haze pollution problem is serious but also the ability to control it is limited. Despite of a certain degree of meteorological factors, in China, haze/smoke events are ultimately arising from inefficient environmental regulations (ERS), unbalanced industry structures, as well as irrational energy structures.
In March 2017, Premier Li first mentioned "fighting against the blue skies" at the fifth meeting of the Twelfth National People's Congress, especially emphasized the importance of strict environmental enforcement and supervision. It shows the control of environmental pollution depends largely on the intensity, supervision and execution of ERS. Although the haze's cross-border nature and spillover characteristics make the management of haze complex, the incidents of "APEC Blue" and "Olympic Blue" have fully demonstrated that ERS is very effective in a special period. Therefore, ERS should be regarded as a necessary mean in addressing the serious air pollution.
ERS can be divided into formal regulation and informal regulation, including market-based tools and command-control tools [3]. Zeyi Zhang and Baoliang Xu (2017) [4] researched the effects of ERS on environmental pollution with introducing hidden economy. Based on the mediating effect method,  [5] analyzed the influence path of ERS on haze pollution and its heterogeneity. Shubin Wang and Yingzhi Xu (2015) [6] studied the decoupling effect of ERS and haze pollution from the perspective of enterprise investment preferences. Danhe Liu and Xiaochen Wang (2017) [3] reviewed studies on ERS theory, including the game of ERS based on different political systems and market mechanisms, and the effectiveness of ERS tools and regulation policies. While many studies have investigated the effects, paths, heterogeneities, policies, mechanisms of ERS, its spatial analysis in china is largely neglected.

Data sources
To analyze ERS, the panel data of 31 provinces in China during the period of 2005-2015 is used in this study. All the original economic and regulated data are taken from National Statistical Yearbook and China Statistical Yearbook for Regional Economy.

Exploratory spatial data analysis (ESDA)
The technology of ESDA is an effective method to analyze the spatial spillover effect of environmental regulations [7], including global spatial auto-dependence (GSA) and local spatial autodependence (LSA).

Global spatial auto-dependence.
Usually, GSA is applied to describe spatial distribution characteristics in the entire study area and is measured by the indices of global ' Moran s I , as: Where ' Moran s I (values between -1 and 1) reflects the degree of similarity of attribute values of each neighbouring spatial regions. Further, n presents the 31 provinces and regions in mainland China. i x And j x are observed annual averaged ERS concentrations from regions i and j , respectively. ij w Is a spatial weight matrix. x Is the average observed variables in different regions while 2 S is the corresponding variance. Standardize ' Moran s I as:

ERS Measurement
The measurement for ERS is challenging due to the large number of missing and unavailable data in the field of air pollution. In the existing literature, there are several common approaches, including concentrations or emissions [8], per capita GDP [9], industry's annual operating cost associated with pollution control [10], pollution abatement control expenditures (PACE) [11], pollution abatement fees and pollution discharge fees [12]. Obviously, the shortcoming of above methods is the single indicator can only reflect a certain aspect of ERS, which can be problematic in our case. Therefore, in order to accurately measure the intensity of ERS in various regions of China, in this paper, we use comprehensive index assessment method (CIAM) to estimate ERS. The system of CIAM consists of one target layer (ERS composite index), three evaluation index layers (waste water, waste gas and solid waste), and a number of individual indicator layers. The calculation steps are as below: First

Global Spatial Correlation Analysis
The results of standardized global Moran's I values are listed in Table 1. Except for the year of 2014, Global Moran's I values of ERS are generally above 0.1 and pass the significant test of p<0.05, indicating the significant positive correlation of spatial distribution for ERS in China rather than distributing randomly. In addition, ERS put up a spatial agglomeration phenomenon in some regions, that is, a high ERS region is often adjacent to a high ERS region, and vice versa. Meanwhile, global Moran's I values vary from 0.1104 to 0.2145, and higher value means better agglomeration effect.

Local Spatial Correlation Analysis
To get a better understanding of spatial characteristics of ERS in China, the local spatial correlation is analyzed in this study including MSP and LISA.

Conclusions
This research studies spatial distribution and spatial correlation of ERS in China by using ESDA method. Results indicate that high ERS regions are expending to Beijing-Tianjin-Hebei region due to the different spatial distribution of ERS in every year. Global spatial correlation analysis shows that ERS in China from 2005 to 2015 exists significant positive correlation, which illustrates that high ERS regions tend to be adjacent to high ERS regions and vice versa. In addition, agglomeration phenomenon is obvious and stable, where High-High aggregation type is mainly distributed in north China, and low-low, low-high and high-low concentration areas are not appeared during the study year. Nowadays, China can no longer avoid the air pollution problem along with its getting worse, Chinese government is keen to tackle the detrimental health impact from air pollution through implementing a policy of environmental regulations. Findings in this study could provide reference for the formulation of ERS governance policies.