Using Machine Learning in Environmental Tax Reform Assessment for Sustainable Development: A Case Study of Hubei Province, China
<p>Twelve Target Cities in Hubei Province in central China. Notes: The gray parts are the pilot cities in this paper.</p> "> Figure 2
<p>Trends in SO<sub>2</sub> Emissions Intensity: Target Cities versus Synthetic Cities (Wuhan, Xiangyang, Shiyan and Ezhou). Notes: Solid lines represent the target cities. Dash lines correspond to the synthetic cities. Vertical dash lines delineate time of reform. (<b>a</b>–<b>d</b>) Wuhan, Xiangyang, Shiyan and Ezhou, respectively, all of which have good fitting and obvious emission intensity reduction effects. Source: authors’ computation by synthetic control program in STATA 12.</p> "> Figure 3
<p>Trends in SO<sub>2</sub> Emissions Intensity: Target Cities versus Synthetic Cities (Jingzhou, Jingmen, Xiaogan and Xianning). Notes: Solid lines represent the target cities. Dash lines correspond to the synthetic cities. Vertical dash lines delineate time of reform. (<b>a</b>–<b>d</b>) Jingmen, Jingzhou, Xiaogan and Xianning, respectively, all of which have good fitting but no great emission intensity reduction effects. Source: authors’ computation by synthetic control program in STATA 12.</p> "> Figure 4
<p>Trends in SO<sub>2</sub> Emissions Intensity: Target Cities versus Synthetic Cities (Yichang, Suizhou, Huanggang and Huangshi). Notes: Solid lines represent the target cities. Dash lines correspond to the synthetic cities. Vertical dash lines delineate time of reform. (<b>a</b>–<b>d</b>) Yichang, Suizhou, Huanggang and Huangshi, respectively, none of which has good fitting. Source: authors’ computation by synthetic control program in STATA 12.</p> "> Figure 5
<p>Ratio Distributions of Post-reform RMSPE to Pre-reform RMSPE: Target Cities and Control Cities. Notes: We remove bad fitting placebo results whose Pre-reform RMSPE exceeds three times the overall RMSPE, and the rest of results are made into a frequency histogram, in which a gray bar represents target cities falling in that interval. Source: authors’ computation by synthetic control program in STATA 12.</p> "> Figure 6
<p>Leave-One-Out Distributions of the Synthetic Control for Target Cities (Wuhan, Xiangyang, Shiyan and Ezhou, <b>a</b>–<b>d</b>). Notes: Solid lines represent the target cities. Dash lines correspond to the synthetic cities. Vertical dash lines delineate time of reform. Each dot-dashed line represents the trend of a new synthetic city’s SO<sub>2</sub> Emissions Intensity, after removing a city that contributes to the previous synthetic city from control group. This figure shows whether the EFT reform effects are affected by synthetic cities’ weights or missing a certain control group city. Source: authors’ computation by synthetic control program in STATA 12.</p> "> Figure 7
<p>Temporal and spatial changes of the intensity rise of SO<sub>2</sub> emission intensity reductions in Hubei Province during 2007–2013.</p> ">
Abstract
:1. Introduction
2. Policy Background
2.1. Background of Hubei Province EFT
2.2. Current EFT Pilot Tax System
3. Method and Data
3.1. Synthetic Control Method and Machining Learning
- (1)
- Policy endogeneity, which means that there is a systematic difference between pilot cities and other cities, and this difference is exactly the reason for why the cities to become pilot cities;
- (2)
- The time and regional restrictions of the pilot areas cannot provide a large number of policy implementation samples; this is not applicable to the traditional econometric methods. Synthetic control method provides a new approach to identifying the effectiveness of the policy. This approach conducts classification and clustering learning according to the characteristics of a large number of non-pilot cities through machine learning, and extracts the core elements to establish a multidimensional model image of pilot cities. By means of linkage of small samples of pilot and big data, the effective assessment of performance on the pilot cities is implemented.
3.2. Data and Sample Selection
- (a)
- Economic factors: Grossman and Krueger [45] first discovered inverted U shape curve of relationship between the degree of environmental pollution and per capita GDP, during the research on the environmental effect of North American Free Trade Agreement, which was later called “Environmental Kuznets Curve” (EKC). Shen [46] found that China’s SO2 emissions are also in line with the characteristics of EKC. In this paper, we use per capita GDP to measure economic development level, recorded as GDP.
- (b)
- Population factors: Urbanization promotes population concentration, which may lead to the scale effect of energy use and reduce pollution emissions [46,47], while the increase of per capita emissions may appear in sparsely populated areas due to the reduction of environmental regulations [48]. In addition, Wang [21] found that people with higher educational degrees to have a stronger awareness of environmental protection, and also to actively fight against pollution. This paper uses the population density, which is the ratio of population at the end of a year and the area of the administrative area, to measure population concentration, and the education population ratio, which is the ratio of number of college students and population at the end of a year, to measure education level. The two variables are recorded as population (pop, thereafter) and education (edu, thereafter) respectively.
- (c)
- Factors of openness: In accordance with the principle of comparative advantage, the pollution intensive industries will be transferred from developed countries to developing countries, or from countries with strong environmental regulations to control weak countries. This is called Pollution Haven Hypothesis [49]. Research [50,51] found that the opening up to foreign investment and the trade openness [52] are related to the SO2 pollution. This paper uses the ratio of foreign industrial output and industrial output to measure the degree of economic openness, recorded as open.
- (d)
- Administrative factors: Wang et al. [53] pointed out that the Chinese enterprises have a strong bargaining power with local environmental protection agencies, resulting in environmental legislation not being fully implemented. The will and enforcement of environmental protection agencies have a great influence on the actual effects of policies. This paper uses the local fiscal dependence on industries, that is, the ratio of VAT payable of Industrial Enterprises above Designated Size and budgetary revenue for measurement, recorded as dependence (dep, thereafter).
- (e)
- Technological factors: Progress of environmental protection technologies and R&D investment will directly improve the environmental protection effects. [46] This paper uses the ratio of expenditure on science and technology to government’s budgeted expenditures for measurement, recorded as technologies (tech, thereafter).
- (f)
- Industrial structure factors: EKC curve shows that environmental pollution is closely related to transformation of industrial structure [54]. Industrialization in the early stage of economic development often leads to rapid increase of natural resources consumption and waste emissions. At later stages of development, the proportion of service industry normally increases, reducing the dependence on the exploitation of resources and energy consumption. The environmental pressure of industrial production is also reduced by technological and management innovation. In this paper, the environmental effect of industrial structure adjustment is measured by the ratio of second industry GDP to the tertiary industry GDP, recorded as second industry (seci, thereafter) and tertiary industry (teri, thereafter).
3.3. Enterprise Interview
4. Results
4.1. Constructing a Synthetic Version of City
4.2. Placebo Studies
4.3. Robustness Evaluation
4.4. Spatial and Temporal Variation of EFT Effect
4.5. Results of Enterprise Interview
5. Discussion
6. Conclusions
Acknowledgments
Author Contributions
Conflicts of Interest
References
- Zhang, S. Environmental Tax: Arrows on the String with Right Target?—A Discussion on Some Key Issues for Environmental Taxation Policy in China. Environ. Prot. 2015, 43, 31–35. [Google Scholar]
- Ge, C.; Gao, S.; Ren, Y.; Sun, G.; Long, F. Study on the design of wastewater environmental tax in China: From wastewater pollutant discharge fee to environmental tax. In Environmental Taxation in China and Asia-Pacific: Achieving Environmental Sustainability through Fiscal Policy; Edward Elgar: Cheltenham, UK, 2011. [Google Scholar]
- Brys, B.; Matthews, S.; Herd, R.; Wang, X. Tax Policy and Tax Reform in the People's Republic of China; OECD Publishing: Paris, France, 2013. [Google Scholar]
- Brys, B.; Matthews, S.; Owens, J. Tax Reform Trends in OECD Countries; OECD Publishing: Paris, France, 2011. [Google Scholar]
- Hoerner, J.; Bosquet, B. Environmental Tax Reform: The European Experience; Center for a Sustainable Economy: Washington, DC, USA, 2001. [Google Scholar]
- Ekins, P.; Speck, S. Environmental Tax Reform (ETR); Oxford University Press: Oxford, UK, 2011. [Google Scholar]
- Withana, S.; Brink, P.; Illes, A.; Nanni, S.; Watkins, E. Environmental tax reform in Europe: Opportunities for the future final report. J. Environ. Econ. Manag. 2005, 49, 174–196. [Google Scholar]
- Harrison, K. A tale of two taxes: The fate of environmental tax reform in Canada. Rev. Policy Res. 2012, 29, 383–407. [Google Scholar] [CrossRef]
- Seung-Joon, P.; Ogawa, Y.; Kawakatsu, T.; Pollitt, H. The double dividend of an environmental tax reform in East Asian economies. In Low-Carbon, Sustainable Future in East Asia: Improving Energy Systems; Lee, S., Pollitt, H., Park, S., Eds.; Taxation and Policy Cooperation: London, UK, 2009. [Google Scholar]
- Ciaschini, M.; Pretaroli, R.; Severini, F.; Socci, C. Regional double dividend from environmental tax reform: An application for the Italian economy. Res. Econ. 2012, 66, 273–283. [Google Scholar] [CrossRef] [Green Version]
- Parry, I.; Norregaard, J.; Heine, D. Environmental tax reform: Principles from theory and practice. Annu. Rev. Resour. Econ. 2012, 4, 101–125. [Google Scholar] [CrossRef]
- Ekins, P.; Pollitt, H.; Barton, J.; Blobel, D. The implications for households of environmental tax reform (ETR) in Europe. Ecol. Econ. 2011, 70, 2472–2485. [Google Scholar] [CrossRef]
- Ekins, P.; Summerton, P.; Thoung, C.; Lee, D. A major environmental tax reform for the UK: Results for the economy, employment and the environment. Environ. Resour. Econ. 2011, 50, 447–474. [Google Scholar] [CrossRef]
- Ekins, P.; Pollitt, H.; Summerton, P.; Chewpreecha, U. Increasing carbon and material productivity through environmental tax reform. Energy Policy 2012, 42, 365–376. [Google Scholar] [CrossRef]
- Barker, T.; Junankar, S.; Pollitt, H.; Summerton, P. The macroeconomic effects of unilateral environmental tax reforms in Europe, 1995–2012. In Critical Issues in Environmental Taxation 1995–2005; Oxford University Press: Oxford, UK, 2009. [Google Scholar]
- Lutz, C.; Meyer, B. Environmental and economic effects of post-Kyoto carbon regimes: Results of simulations with the global model GINFORS. Energy Policy 2009, 37, 1758–1766. [Google Scholar] [CrossRef]
- Barker, T.; Junankar, S.; Pollitt, H.; Summerton, P. The effects of environmental tax reform on international competitiveness in the European Union: Modelling with E3ME. In Carbon-Energy Taxation: Lessons from Europe; Andersen, M., Ekins, P., Eds.; Oxford University Press: Oxford, UK, 2009. [Google Scholar]
- Dasgupta, S.; Huq, M.; Wheeler, D.; Zhang, C. Water pollution abatement by Chinese industry: Cost estimates and policy implications. Appl. Econ. 2001, 33, 547–557. [Google Scholar] [CrossRef]
- Wang, H.; Wheeler, D. Pricing Industrial Pollution in China: An Econometric Analysis of the Levy System; World Bank Publications: Washington, DC, USA, 1996. [Google Scholar]
- Wang, H.; Wheeler, D. Endogenous Enforcement and Effectiveness of China’s Pollution Levy System; World Bank Publications: Washington, DC, USA, 2000. [Google Scholar]
- Wang, H.; Wheeler, D. Financial incentives and endogenous enforcement in China’s pollution levy system. J. Environ. Econ. Manag. 2005, 49, 174–196. [Google Scholar] [CrossRef]
- Chen, S. Marginal abatement cost and environmental tax reform in China. Soc. Sci. China 2011, 3, 6–15. [Google Scholar] [CrossRef]
- Tong, J.; Zhu, B. A Study on “Environmental Dividend” as the Primary Objective in Environmental Tax Reform in China. Int. Tax. China 2011, 8, 11–21. [Google Scholar]
- Xiao, B.; Niu, D.; Guo, X.; Xu, X. The Impacts of Environmental Tax in China: A Dynamic Recursive Multi-Sector CGE Model. Energies 2015, 8, 7777–7804. [Google Scholar] [CrossRef]
- Liang, Q.; Yao, Y.; Zhao, L.; Yao, Y.; Wang, C.; Yang, R.; Wei, Y. Platform for China Energy and Environmental Policy Analysis: A general design and its application. Environ. Model. Softw. 2014, 51, 195–206. [Google Scholar] [CrossRef]
- Blanchard, O.; Kahn, C. The Solution of Linear Difference Models under Rational Expectations. Econometrica 1980, 48, 1305–1311. [Google Scholar] [CrossRef]
- Puhani, P. The treatment effect, the cross difference, and the interaction term in nonlinear “difference-in-differences” models. Econ. Lett. 2012, 115, 85–87. [Google Scholar] [CrossRef]
- Abadie, A.; Diamond, A.; Hainmueller, J. Synthetic Control Methods for Comparative Case Studies: Estimating the Effect of California’s Tobacco Control Program. J. Am. Stat. Assoc. 2010, 105, 493–505. [Google Scholar] [CrossRef]
- Abadie, A.; Gardeazabal, J. The Economic Costs of Conflict: A Case Study of the Basque Country. Am. Econ. Rev. 2003, 93, 113–132. [Google Scholar] [CrossRef]
- Abadie, A.; Diamond, A.; Hainmueller, J. Synth: An r package for synthetic control methods in comparative case studies. J. Stat. Softw. 2011, 42, 1–17. [Google Scholar] [CrossRef]
- Abadie, A.; Diamond, A.; Hainmueller, J. Comparative Politics and the Synthetic Control Method. Am. J. Polit. Sci. 2015, 59, 495–510. [Google Scholar] [CrossRef]
- Saunders, J.; Lundberg, R.; Braga, A.; Ridgeway, G.; Miles, J. A Synthetic Control Approach to Evaluating Place-Based Crime Interventions. J. Quant. Criminol. 2014, 31, 1–22. [Google Scholar] [CrossRef]
- Sills, E.; Herrera, D.; Kirkpatrick, A.; Brandão, A.; Dickson, R.; Hall, S.; Pattanayak, S.; Shoch, D.; Vedoveto, M.; Young, L.; et al. Estimating the Impacts of Local Policy Innovation: The Synthetic Control Method Applied to Tropical Deforestation. PLoS ONE 2015, 10, e0132590. [Google Scholar] [CrossRef] [PubMed]
- Ando, M. Dreams of urbanization: Quantitative case studies on the local impacts of nuclear power facilities using the synthetic control method. J. Urban Econ. 2015, 85, 68–85. [Google Scholar] [CrossRef]
- Zhang, J.; Zhong, C.; Yi, M. Did Olympic Games improve air quality in Beijing? Based on the synthetic control method. Environ. Econ. Policy Stud. 2015, 18, 21–39. [Google Scholar] [CrossRef]
- Sra, S.; Nowozin, S.; Wright, S.J. Optimization for Machine Learning; Mit Press: Cambridge, UK, 2012. [Google Scholar]
- National Bureau of Statistics of China. China City Statistical Yearbook; Chinese Statistics Press: Beijing, China, 2014.
- China Environment Yearbook Editing Committee. China Environment Yearbook; China Environment Yearbook Press: Beijing, China, 2014. [Google Scholar]
- Peng, Y.; Cui, J.; Cao, Y. Impact of Manufacturing Transfer on SO2 Emissions in Jiangsu Province, China. Atmosphere 2016, 7, 69. [Google Scholar] [CrossRef]
- Lehoczki, Z.; Sleszynski, J. Sulphur Dioxide Emission Charge in Poland. In Economic Instruments for Environmental Management: A Worldwide Compendium of Case Studies; Routledge: London, UK, 2014. [Google Scholar]
- Nam, K.; Waugh, C.; Paltsev, S.; Reilly, J.; Karplus, V. Synergy between pollution and carbon emissions control: Comparing China and the United States. Energy Econ. 2014, 46, 186–201. [Google Scholar] [CrossRef]
- Zhao, Y.; Duan, L.; Larssen, T.; Nielsen, C.; Hao, J. Soil acidification in China: Is controlling SO2 emissions enough? Environ. Sci. Technol. 2009, 43, 8021–8026. [Google Scholar] [CrossRef] [PubMed]
- Xie, Y.; Zhao, B.; Zhang, L.; Luo, R. Spatiotemporal variations of PM2.5 and PM10 concentrations between 31 Chinese cities and their relationships with SO2, NO2, CO and O3. Particuology 2015, 20, 141–149. [Google Scholar] [CrossRef]
- Lai, H. Developing Central China: A new regional programme. China Int. J. 2007, 5, 109–128. [Google Scholar] [CrossRef]
- Grossman, G.; Krueger, A. Environmental Impacts of a North American Free Trade Agreement. 1991. Available online: http://www.nber.org/papers/w3914.pdf?newwindow=l (accessed on 31 October 2016).
- Shen, J. A Simultaneous Estimation of Environmental Kuznets Curve: Evidence from China. China Econ. Rev. 2006, 17, 383–394. [Google Scholar] [CrossRef]
- Liu, Y.; Zhou, Y.; Wu, W. Assessing the impact of population, income and technology on energy consumption and industrial pollutant emissions in China. Appl. Energy 2015, 155, 904–917. [Google Scholar] [CrossRef]
- Selden, T.; Song, D. Environmental quality and development: Is there a Kuznets curve for air pollution emissions? J. Environ. Manag. 1994, 27, 147–162. [Google Scholar] [CrossRef]
- Copeland, B.; Taylor, M. Trade and transboundary pollution. Am. Econ. Rev. 1995, 85, 716–737. [Google Scholar]
- Lan, J.; Kakinaka, M.; Huang, X. Foreign direct investment, human capital and environmental pollution in China. Environ. Res. Econ. 2012, 51, 255–275. [Google Scholar] [CrossRef]
- Bao, Q.; Chen, Y.; Song, L. Foreign direct investment and environmental pollution in China: A simultaneous equations estimation. Environ. Dev. Econ. 2011, 16, 71–92. [Google Scholar] [CrossRef]
- Jayanthakumaran, K.; Liu, Y. Openness and the environmental Kuznets curve: Evidence from China. Econ. Model. 2012, 29, 566–576. [Google Scholar] [CrossRef]
- Wang, H.; Mamingi, N.; Laplante, B.; Dasgupta, S. Incomplete enforcement of pollution regulation: Bargaining power of Chinese factories. Environ. Resour. Econ. 2003, 24, 245–262. [Google Scholar] [CrossRef]
- Dasgupta, S.; Laplante, B.; Wang, H.; Wheeler, D. Confronting the environmental Kuznets curve. J. Econ. Perspect. 2002, 16, 147–168. [Google Scholar] [CrossRef]
- Chen, Y. Research of Environmental Taxation Reform in China. Master’s Thesis, Central China Normal University, Wuhan, China, 2014. [Google Scholar]
- Feng, Y. Research on Green Development & Environmental Tax System Reform in China. Master’s Thesis, Central China Normal University, Wuhan, China, 2014. [Google Scholar]
- Wang, A. The search for sustainable legitimacy: Environmental law and bureaucracy in China. Harv. Environ. Law Rev. 2013, 37, 365–589. [Google Scholar] [CrossRef]
- Costantini, V.; Mazzanti, M.; Montini, A. Environmental performance, innovation and spillovers. Evidence from a regional NAMEA. Ecol. Econ. 2013, 89, 101–114. [Google Scholar] [CrossRef]
Item Compared | Pollutant Discharge Fees | Environmental Taxes |
---|---|---|
Collection department | Environmental protection department of province, city, county | Provincial local tax department |
Collection cycle | Half of year or a year | Month or season |
Funds use | Funds are used for environment governance; however, irregularities led to self-use funds for the environmental protection department or the local government. | Funds are under the unified management of public finance budget. |
Collection procedure | The environmental protection department is responsible for the approval and collection, and the payment is into the local financial budget account. A strict monitoring and evaluation mechanism is absent. | It is approved by the environmental protection department, and collected into the treasury by the tax department under the same tax assessment mechanism. |
Enforcement | Its enforcement is weaker than that of tax. As a soft constraint, its law enforcement is relatively large at random. | As an obligatory constraint, it has law enforcement with rigid features. |
Superior leadership | Local government | Provincial government |
City | Weight |
---|---|
Xian | 0.420 |
Shanghai | 0.132 |
Beijing | 0.125 |
Changsha | 0.078 |
Shantou | 0.066 |
Haikou | 0.064 |
Daqing | 0.057 |
Xiangtan | 0.021 |
Jiayuguan | 0.017 |
Benxi | 0.012 |
Jinzhou | 0.010 |
Wuhan | Synthetic Wuhan | Average of 262 Control Cities | |
---|---|---|---|
Per capita GDP | 2.7628 | 2.8206 | 1.6247 |
seci | 0.4566 | 0.4578 | 0.4706 |
teri | 0.4944 | 0.4950 | 0.3599 |
pop | 945.5 | 945.4 | 414.0 |
dep | 0.1780 | 0.1788 | 0.2091 |
tech | 0.0089 | 0.0089 | 0.0049 |
edu | 0.0822 | 0.0819 | 0.0113 |
open | 2.9601 | 2.9748 | 1.8989 |
SO2 emissions intensity (2003) | 0.0067 | 0.0067 | 0.0187 |
SO2 emissions intensity (2005) | 0.0060 | 0.0060 | 0.0167 |
SO2 emissions intensity (2007) | 0.0041 | 0.0041 | 0.0117 |
City | Average Reduction (Ton/Million Yuan) | Average Decline Rate (%) |
---|---|---|
Wuhan | 0.0329 | 16.82 |
Xiangyang | 0.1280 | 30.21 |
Ezhou | 0.1936 | 19.16 |
Shiyan | 0.1490 | 32.42 |
Xianning | 0.1857 | 34.34 |
Jingzhou | −0.0051 | −3.04 |
Xiaogan | 0.0604 | 10.37 |
Jingmen | 0.0963 | 16.68 |
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zheng, Y.; Zheng, H.; Ye, X. Using Machine Learning in Environmental Tax Reform Assessment for Sustainable Development: A Case Study of Hubei Province, China. Sustainability 2016, 8, 1124. https://doi.org/10.3390/su8111124
Zheng Y, Zheng H, Ye X. Using Machine Learning in Environmental Tax Reform Assessment for Sustainable Development: A Case Study of Hubei Province, China. Sustainability. 2016; 8(11):1124. https://doi.org/10.3390/su8111124
Chicago/Turabian StyleZheng, Yinger, Haixia Zheng, and Xinyue Ye. 2016. "Using Machine Learning in Environmental Tax Reform Assessment for Sustainable Development: A Case Study of Hubei Province, China" Sustainability 8, no. 11: 1124. https://doi.org/10.3390/su8111124