基于产业结构门槛效应模型的环境政策治污效益评估——以《大气污染防治行动计划》为例

doi:10.18402/resci.2021.02.12

摘要/Abstract

摘要：

环境质量的改善不仅是环境政策单独作用的结果,通常还与政策作用下产业结构的调整密切相关。本文基于中国2006—2016年197个地级及以上城市的PM2.5年均浓度面板数据,以“大气十条”政策作为环境规制的代理虚拟变量,结合双重差分法,研究“大气十条”政策的实施是否有利于降低京津冀晋鲁蒙、长三角、珠三角3个目标区域内城市的雾霾污染水平,并通过构建产业结构升级指标,探索该政策如何通过产业结构升级影响雾霾污染治理。进一步,采用面板门槛回归模型,验证“大气十条”政策实施背景下产业结构合理化与产业结构高度化对雾霾污染治理是否存在门槛效应。回归结果显示：①双重差分回归中,全样本下,“大气十条”的政策效果不显著;分区域样本下,政策对京津冀晋鲁蒙区域产生了显著的雾霾治理效应,而对长三角和珠三角区域没有产生明显的政策效果。②中介效应模型中,京津冀晋鲁蒙区域“大气十条”政策主要通过产业结构高度化的“量”作用于雾霾污染治理,且为部分中介效应;长三角区域产业结构高度化的“量”作用于雾霾污染治理则为完全中介效应;珠三角区域产业结构升级不具有显著的中介效应。③面板门槛回归中,京津冀晋鲁蒙与周边区域样本、长三角与周边区域样本下产业结构合理化与高度化均存在门槛效应。

关键词: “大气十条”政策, 环境规制, 雾霾污染, 双重差分法, 门槛效应, 产业结构合理化, 产业结构高度化

Abstract:

The improvement of environmental quality is not only the result of environmental policy, but closely related to the adjustment of industrial structure under the effect of policy. Based on the panel data of the average annual concentration of PM2.5 in 197 prefectural-level and above cities in China from 2006 to 2016, this study took the Air Pollution Prevention and Control Action Plan as the proxy dummy variable of environmental regulation, and combined the difference in differences method to examine whether the implementation of the action plan is conducive to reducing the level of haze pollution in three target areas. By putting the industrial structure upgrading indicators (including industrial structure optimization, the “quantity” of industrial structure supererogation, and the “quality” of industrial structure supererogation) into the model, this study explored how the policy affected the haze pollution control through the upgrading of industrial structure. In addition, a panel threshold regression model was used to examine whether there is a threshold effect of industrial structure optimization and industrial structure supererogation on haze pollution control under the background of the implementation of the action plan. The regression results show that: (1) Using the full sample, the policy effect is not significant. In the subregional samples, the policy has a significant haze control effect in Beijing, Tianjin, Hebei and the surrounding areas, but has no significant policy effect on the Yangtze River Delta and the Pearl River Delta; (2) In the mediation effect model, in Beijing, Tianjin, Hebei and the surrounding areas the policy mainly acts on haze pollution control through the “quantity” of industrial structure supererogation, and the effect is partial intermediary. The “quantity” of industrial structure supererogation in the Yangtze River Delta is complete intermediary when it acts on haze pollution control and the industrial structure upgrading variables in the Pearl River Delta region do not have a significant mediating effect; (3) In the panel threshold regression, threshold effects of industrial structure optimization and industrial structure supererogation are observed in the samples of Beijing-Tianjin-Hebei-Shanxi-Shandong-Inner Mongolia and surrounding areas, and the samples of the Yangtze River Delta and surrounding areas.

Key words: Air Pollution Prevention and Control Action Plan, environmental regulation, haze pollution, difference-in-differences method, threshold effect, industrial structure optimization, industrial structure supererogation

陈林, 肖倩冰, 蓝淑菁. 基于产业结构门槛效应模型的环境政策治污效益评估——以《大气污染防治行动计划》为例[J]. 资源科学, 2021, 43(2): 341-356.

CHEN Lin, XIAO Qianbing, LAN Shujing. Pollution control effects of environmental policies based on threshold effect model of industrial structure: Taking the Air Pollution Prevention and Control Action Plan as an example[J]. Resources Science, 2021, 43(2): 341-356.

图/表 5

表1

表2

表3

表4

图1

参考文献 45

[1]	World Bank. World Development Indicators 2007[R]. Washington: The World Bank, 2007.
[2]	Zhang Q, Crooks R. Toward an Environmentally Sustainable Future: Country Environmental Analysis of the People’s Republic of China[R]. ADB Reports RPT113974, 2012.
[3]	罗知, 李浩然. “大气十条” 政策的实施对空气质量的影响[J]. 中国工业经济, 2018, (9):136-154.
	[Luo Z, Li H R. The impact of “Atmosphere Ten Articles” policy on air quality in China[J]. China Industrial Economics, 2018, (9):136-154.]
[4]	张扬, 付淩波, 李薇, 等. 基于黑龙江省大气污染防治行动计划的温室气体减排核算[J]. 中国人口·资源与环境, 2015,25(S2):333-336.
	[Zhang Y, Fu L B, Li W, et al. Accounting for the reduction of greenhouse gas emissions based on the air pollution control action plan in Heilongjiang Province[J]. China Population, Resources and Environment, 2015,25(S2):333-336.]
[5]	曹颖. 大气污染应从末端治理转向系统解决: 兼谈《大气污染防治行动计划》对能源结构的影响[J]. 宏观经济管理, 2014, (4):41-43.
	[Cao Y. Air pollution should shift from terminal treatment to system solution: Discussion on the impact of “Air Pollution Prevention and Control Action Plan”on energy structure[J]. Macroeconomic Management, 2014, (4):41-43.]
[6]	马国霞, 周颖, 吴春生, 等. 成渝地区《大气污染防治行动计划》实施的成本效益评估[J]. 中国环境管理, 2019,11(6):38-43.
	[Ma G X, Zhou Y, Wu C S, et al. Cost-benefit assessment of impacts of China’s National Air Pollution Action Plan in Cheng-Yu Region[J]. Chinese Journal of Environmental Management, 2019,11(6):38-43.]
[7]	易兰, 周忆南, 李朝鹏, 等. 城市机动车限行政策对雾霾污染治理的成效分析[J]. 中国人口·资源与环境, 2018,28(10):81-87.
	[Yi L, Zhou Y N, Li Z P, et al. Analysis of the effects of driving restriction policies in controlling haze pollution[J]. China Population, Resources and Environment, 2018,28(10):81-87.]
[8]	沈洪涛, 周艳坤. 环境执法监督与企业环境绩效: 来自环保约谈的准自然实验证据[J]. 南开管理评论, 2017,20(6):73-82.
	[Shen H T, Zhou Y K. Supervision of environmental policy enforcement and firm environmental performance: Evidence from a Quasi-natural experiment[J]. NanKai Business Review, 2017,20(6):73-82.]
[9]	孙振清, 李欢欢, 刘保留. 碳交易政策下区域减排潜力研究: 产业结构调整与技术创新双重视角[J]. 科技进步与对策, 2020,37(5):28-35.
	[Sun Z Q, Li H H, Liu B L. Research on regional emission reduction potential under carbon trading policy: The perspective of industrial restructuring and technological innovation[J]. Science & Technology Progress and Policy, 2020,37(5):28-35.]
[10]	杨骞, 秦文晋, 刘华军. 环境规制促进产业结构优化升级吗?[J]. 上海经济研究, 2019, (6):83-95.
	[Zhang Q, Qin W J, Liu H J. Does environmental regulation promote the industrial structure upgrading?[J]. Shanghai Journal of Economics, 2019, (6):83-95.]
[11]	胡晖, 朱钰琦, 方德斌, 等. 环境规制影响产业结构的路径与机制: 基于湘鄂赣皖地区城市的实证研究[J]. 长江流域资源与环境, 2020,29(12):2620-2635.
	[Hu H, Zhu Y Q, Fang D B, et al. The path and mechanism of environmental regulation affecting industrial structure: An empirical study of cities in Hunan, Hubei, Jiangxi and Anhui[J]. Resources and Environment in the Yangtze Basin, 2020,29(12):2620-2635.]
[12]	李虹, 邹庆. 环境规制、资源禀赋与城市产业转型研究: 基于资源型城市与非资源型城市的对比分析[J]. 经济研究, 2018,53(11):182-198.
	[Li H, Zou Q. Environmental regulations, resource endowments and urban industry transformation: Comparative analysis of resource-based and non-resource-based cities[J]. Economic Research Journal, 2018,53(11):182-198.]
[13]	毛建辉, 管超. 环境规制抑制产业结构升级吗? 基于政府行为的非线性门槛模型分析[J]. 财贸研究, 2020,31(3):29-42.
	[Mao J H, Guan C. Does environmental regulation inhibit industrial structure upgrading? Analysis of nonlinear threshold model based on government behavior[J]. Finance and Trade Research, 2020,31(3):29-42.]
[14]	徐志伟. 工业经济发展、环境规制强度与污染减排效果: 基于“先污染, 后治理”发展模式的理论分析与实证检验[J]. 财经研究, 2016,42(3):134-144.
	[Xu Z W. Industrial economic development, environmental regulation intensity and emissions reduction effect: Theoretical analysis and empirical test based on development pattern of “treatment after pollution”[J]. Journal of Finance and Economics, 2016,42(3):134-144.]
[15]	何为, 刘昌义, 刘杰, 等. 环境规制、技术进步与大气环境质量: 基于天津市面板数据实证分析[J]. 科学学与科学技术管理, 2015,36(5):51-61.
	[He W, Liu C Y, Liu J, et al. Environmental regulation, technology change, and air pollution: A panel study on Tianjin[J]. Science of Science and Management of S & T, 2015,36(5):51-61.]
[16]	王红梅. 中国环境规制政策工具的比较与选择: 基于贝叶斯模型平均(BMA)方法的实证研究[J]. 中国人口·资源与环境, 2016,26(9):132-138.
	[Wang H M. Comparison and selection of environmental regulation policy in China: Based on Bayesian model averaging approach[J]. China population, Resources and Environment, 2016,26(9):132-138.]
[17]	申晨, 贾妮莎, 李炫榆. 环境规制与工业绿色全要素生产率: 基于命令控制型与市场激励型规制工具的实证分析[J]. 研究与发展管理, 2017,29(2):144-154.
	[Shen C, Jia N S, Li X Y. Environmental regulation and industrial green total factor productivity: Empirical analysis based on CAC and MBI environmental regulations[J]. R&D Management, 2017,29(2):144-154.]
[18]	李强. 正式与非正式环境规制的减排效应研究: 以长江经济带为例[J]. 现代经济探讨, 2018, (5):92-99.
	[Li Q. Research on the emission reduction effects of formal and informal environmental regulations: Taking the Yangtze River Economic Belt as an example[J]. Modern Economic Research, 2018, (5):92-99.]
[19]	张友国. 经济发展方式变化对中国碳排放强度的影响[J]. 经济研究, 2010,45(4):120-133.
	[Zhang Y G. Economic development pattern change impact on China’s carbon intensity[J]. Economic Research Journal, 2010,45(4):120-133.]
[20]	Jalil A, Feridun M. The impact of growth, energy and financial development on the environment in China: A cointegration analysis[J]. Energy Economics, 2011,33(2):284-291. doi: 10.1016/j.eneco.2010.10.003
[21]	李斌, 赵新华. 经济结构、技术进步与环境污染: 基于中国工业行业数据的分析[J]. 财经研究, 2011,37(4):112-122.
	[Li B, Zhao X H. Economic structure, technological progress and environmental pollution: Based on the analysis of industrial data in China[J]. Journal of Finance and Economics, 2011,37(4):112-122.]
[22]	杨志明, 鄢哲明. 污染治理视角下中国城市工业环境库兹涅茨曲线假说的再检验[J]. 浙江社会科学, 2013(6):31-39.
	[Yang Z M, Yan Z M. The Re-examination of the environmental Kuznets Curve Hypojournal of Chinese urban industrial environment from the perspective of pollution control[J]. Zhejiang Social Sciences, 2013(6):31-39.]
[23]	孙坤鑫, 钟茂初. 环境规制、产业结构优化与城市空气质量[J]. 中南财经政法大学学报, 2017, (6):63-72.
	[Sun K X, Zhong M C. Environmental regulation, industrial structure optimization and urban air quality[J]. Journal of Zhongnan University of Economics and Law, 2017, (6):63-72.]
[24]	刘晨跃, 徐盈之. 环境规制如何影响雾霾污染治理: 基于中介效应的实证研究[J]. 中国地质大学学报(社会科学版), 2017,17(6):41-53.
	[Liu C Y, Xu Y Z. How does environmental regulation affect the governance of haze pollution? An empirical study based on the mediating effect method[J]. Journal of China University of Geosciences (Social Sciences Edition), 2017,17(6):41-53.]
[25]	屈小娥. 异质型环境规制影响雾霾污染的双重效应[J]. 当代经济科学, 2018,40(6):26-37.
	[Qu X E. Double effects of heterogeneous environmental regulation on haze pollution[J]. Modern Economic Science, 2018,40(6):26-37.]
[26]	杨斯悦, 王凤, 刘娜. 《大气污染防治行动计划》实施效果评估: 双重差分法[J]. 中国人口·资源与环境, 2020,30(5):110-117.
	[Yang S Y, Wang F, Liu N. Assessment of the Air Pollution Prevention and Control Action Plan in China: A difference-in-difference analysis[J]. China population, Resources and Environment, 2020,30(5):110-117.]
[27]	陈诗一, 陈登科. 雾霾污染、政府治理与经济高质量发展[J]. 经济研究, 2018,53(2):20-34.
	[Chen S Y, Chen D K. Air pollution, government regulations and high-quality economic development[J]. Economic Research Journal, 2018,53(2):20-34.]
[28]	牟雪洁, 饶胜, 张箫, 等. 产业发展与生态承载力一致性评价的理论与技术框架构建[J]. 生态经济, 2020,36(3):45-50.
	[Mou X J, Rao S, Zhang X, et al. Construction of theory and technology framework for consistency evaluation of industrial development and ecological carrying capacity[J]. Ecological Economy, 2020,36(3):45-50.]
[29]	Donkelaar A V, Martin R V, Brauer M, et al. Global fine particulate matter concentrations from satellite for long-term exposure assessment[J]. Environment Health Perspectives, 2015,123:135-143. doi: 10.1289/ehp.1408646
[30]	干春晖, 郑若谷, 余典范. 中国产业结构变迁对经济增长和波动的影响[J]. 经济研究, 2011,46(5):4-16.
	[Gan C H, Zheng R G, Yu D F. An empirical study on the effects of industrial structure on economic growth and fluctuations in China[J]. Economic Research Journal, 2011,46(5):4-16.]
[31]	唐宇娣, 朱道林, 程建, 等. 差别定价的产业用地供应策略对产业结构升级的影响: 基于中国277个城市的实证分析[J]. 资源科学, 2020,42(3):548-557.
	[Tang Y D, Zhu D L, Cheng J, et al. Impact of differential pricing strategy of land supply on the upgrading of industrial structure: A study based on the empirical analysis of 277 cities in China[J]. Resources Science, 2020,42(3):548-557.]
[32]	袁航, 朱承亮. 国家高新区推动了中国产业结构转型升级吗[J]. 中国工业经济, 2018, (8):60-77.
	[Yuan H, Zhu C L. Do national high-tech zones promote the transformation and upgrading of China’s industrial structure[J]. China Industrial Economics, 2018, (8):60-77.]
[33]	陈林, 万攀兵. 《京都议定书》及其清洁发展机制的减排效应: 基于中国参与全球环境治理微观项目数据的分析[J]. 经济研究, 2019,54(3):55-71.
	[Chen L, Wan P B. The Kyoto Protocol and its clean development mechanism’s emission reduction effects: Micro-project data on China’s participation in global environmental governance[J]. Economic Research Journal, 2019,54(3):55-71.]
[34]	Khandker S B, Koolwal G, Samad H. Handbook on Impact Evaluation: Quantitative Methods and Practices[M]. Washington: The World Bank, 2010.
[35]	Moser P, Voena A. Compulsory licensing: Evidence from the trading with the enemy act[J]. American Economic Review, 2009,102(1):396-427.
[36]	温忠麟, 叶宝娟. 中介效应分析: 方法和模型发展[J]. 心理科学进展, 2014,22(5):731-745. doi: 10.3724/SP.J.1042.2014.00731
	[Wen Z L, Ye B J. Analyses of mediating effects: The development of methods and models[J]. Advances in Psychological Science, 2014,22(5):731-745.]
[37]	宋雪珺, 王多多, 覃飞, 等. 长三角城市群2010年生态足迹与生态承载力分析[J]. 生态科学, 2018,37(2):162-172.
	[Song X J, Wang D D, Qin F, et al. Analysis of ecological footprint and ecological carrying capacity of Yangtze River Delta urban agglomeration in 2010[J]. Ecological Science, 2018,37(2):162-172.]
[38]	钟茂初. 如何表征区域生态承载力与生态环境质量? 兼论以胡焕庸线生态承载力涵义重新划分东中西部[J]. 中国地质大学学报(社会科学版), 2016,16(1):1-9.
	[Zhong M C. How to characterize the regional ecological carrying capacity and the quality of the ecological environment? Also on the re-division of the eastern, central and western regions with the meaning of the ecological carrying capacity of the Hu Huanyong Line[J]. Journal of China University of Geosciences (Social Sciences Edition), 2016,16(1):1-9.]
[39]	王民, 李秀霞. 基于生态足迹的区域产业结构优化研究[J]. 学术交流, 2009, (2):70-73.
	[Wang M, Li X X. Research on the optimization of regional industrial structure based on ecological footprint[J]. Academic Exchanges, 2009, (2):70-73.]
[40]	刘成坤, 赵昕东. 人口老龄化对中国产业结构升级的影响研究[J]. 区域经济评论, 2018, (4):69-80.
	[Liu C K, Zhao X D. Influence of population aging on the China’s upgrading of industrial structure[J]. Regional Economic Review, 2018, (4):69-80.]
[41]	Hansen B E. Threshold effects in non-dynamic panels: Estimation, testing, and inference[J]. Journal of Econometrics, 1999,93(2):345-368. doi: 10.1016/S0304-4076(99)00025-1
[42]	熊爱华, 丁友强, 胡玉凤. 低碳门槛下绿色创新补贴对全要素生产率的影响[J]. 资源科学, 2020,42(11):2184-2195.
	[Xiong A H, Ding Y Q, Hu Y F. Impact of low-carbon subsidies and green innovation on total factor productivity in view of the threshold effect of carbon emission reduction[J]. Resources Science, 2020,42(11):2184-2195.]
[43]	郭越, 黄友星, 杨宇, 等. 外商直接投资对小岛屿国家经济增长的影响: 基于旅游业门槛效应的分析[J]. 资源科学, 2020,42(11):2132-2144.
	[Guo Y, Huang Y X, Yang Y, et al. Impact of foreign direct investment on Small Island States’ economic growth: An analysis based on the threshold effect of tourism[J]. Resources Science, 2020,42(11):2132-2144.]
[44]	成淑敏, 高阳, 黄姣, 等. 京津冀及江浙沪经济圈生态足迹比较分析[J]. 长江流域资源与环境, 2012,21(4):433-441.
	[Cheng S M, Gao Y, Huang J, et al. Analysis of ecological footprint and ecological carrying capacity: A case study of Jing-Jin-Ji and Jiang-Zhe-Hu region[J]. Resources and Environment in the Yangtze Basin, 2012,21(4):433-441.]
[45]	邬娜, 傅泽强, 谢园园, 等. 基于生态承载力的产业布局优化研究进展述评[J]. 生态经济, 2015,31(5):21-25.
	[Wu N, Fu Z Q, Xie Y Y, et al. Review on the research progress of industrial layout optimization based on ecological carrying capacity[J]. Ecological Economy, 2015,31(5):21-25.]

变量	样本量	平均值	标准差	最小值	最大值
lnPM2.5	2167	3.4938	0.5062	1.5425	4.5093
lnPM2.5_	2167	3.6466	0.4156	1.8961	4.6074
ris	2167	0.2602	0.2060	0.0001	1.7219
ois1	2167	2.2429	0.1400	1.8312	2.7972
ois2	2167	0.9960	0.4770	0.1686	8.0673
lnpopden	2167	2.4696	0.4051	0.6721	3.4229
lngdppc	2167	10.1638	0.7560	8.2013	12.9535
lngdppc2	2167	103.8736	15.6048	67.2622	167.7935
lninvest	2167	2.5395	0.4866	1.5194	4.9021
scale	2167	0.1522	0.0789	0.0118	1.4852
tech	2167	0.0142	0.0136	0.0003	0.1656
open	2167	0.0196	0.0190	1.23E-05	0.1316
temp	2167	14.7718	5.3999	-2.1000	26.4000
lnra	2167	4.6046	0.4524	2.8332	5.5174
humi	2167	65.9720	10.3891	14.9000	95.0000

解释变量	全样本		京津冀晋鲁蒙与周边区域		长三角与周边区域		珠三角与周边区域
解释变量	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
policy	0.007	0.018	-0.092***	-0.072***	0.038***	-0.003	0.000	0.041
	-0.016	(0.014)	(0.025)	(0.023)	(0.013)	(0.013)	(0.014)	(0.028)
经济因素	Y	Y	Y	Y	Y	Y	Y	Y
气象因素	Y	Y	Y	Y	Y	Y	Y	Y
ytreat		-0.002 (0.002)		-0.004 (0.003)		0.009*** (0.002)		-0.009 (0.005)
时间固定效应	Y	-0.002 (0.002)	Y	-0.004 (0.003)	Y	0.009*** (0.002)	Y	-0.009 (0.005)	Y	Y	Y	Y
城市固定效应	Y	Y	Y	Y	Y	Y	Y	Y
Constant	8.350***	10.397***	10.543***	14.813***	63.198***	-3.715	5.746***	9.022***
	(1.245)	(2.715)	(2.659)	(4.712)	(1.463)	(3.604)	(1.342)	(2.476)
Observations	2167	2167	990	990	726	726	561	561
R²	0.262	0.264	0.366	0.367	0.580	0.591	0.741	0.746

样本区间	变量	（1）	（2）	（3）	（4）	（5）
京津冀晋鲁蒙与周边区域	policy	-0.092***				-0.088***
京津冀晋鲁蒙与周边区域		(0.025)				(0.025)
	ris		0.020			-0.148
			(0.021)			(0.089)
	ois1			0.009		-0.718***
				(0.009)		(0.175)
	ois2				0.129**	0.044**
					(0.060)	(0.021)
	趋势项	N				N
长三角与周边区域	policy	-0.004				-0.003
		(0.013)				(0.013)
	ris		0.029			0.052
			(0.019)			(0.047)
	ois1			0.018**		-0.273***
				(0.008)		(0.087)
	ois2				-0.038	-0.026**
					(0.056)	(0.013)
	趋势项	Y				Y
珠三角与周边区域	policy	0.000				-0.002
		(0.014)				(0.015)
	ris		0.014			0.056
			(0.025)			(0.055)
	ois1			-0.001		-0.109
				(0.012)		(0.127)
	ois2				-0.063	-0.017
					(0.042)	(0.019)
	趋势项	N				N

解释变量	京津冀晋鲁蒙与周边区域				解释变量	长三角与周边区域
解释变量	(1)		(2)	(4)	解释变量	(1)	(2)	(4)
policy_n		-0.099***	-0.088***	-0.100***	policy_c	0.043***	0.058***	0.046***
		(0.015)	(0.015)	(0.014)		(0.018)	(0.018)	(0.018)
经济因素		Y	Y	Y	经济因素	Y	Y	Y
气象因素		Y	Y	Y	气象因素	Y	Y	Y
时间趋势		N	N	N	时间趋势	Y	Y	Y
时间固定效应		Y	Y	Y	时间固定效应	Y	Y	Y
城市固定效应		Y	Y	Y	城市固定效应	Y	Y	Y
ris·I (ris≤ 0.445)		-0.053			ris·I (ris≤ 0.409)	0.127**
						(0.062)
ris·I (ris>0.445)		0.059			ris·I (0.409<ris≤ 0.669)	0.013
		(0.046)				(0.052)
					ris·I (0.669<ris≤ 0.762)	0.124**
						(0.056)
					ris·I (ris>0.762)	0.015
						(0.057)
ois1·I (ois1≤ 2.059)			-0.290**		ois1·I (ois1≤ 2.405)		-0.423***
			(0.122)				(0.113)
ois1·I (2.059<ois1≤ 2.535)			-0.323***		ois1·I (ois1>2.405)		-0.448***
			(0.117)				(0.112)
ois1·I (ois1>2.535)			-0.360***
			(0.117)
ois2·I (ois2≤ 0.277)				-0.019	ois2·I (ois2≤ 0.889)			-0.035
				(0.256)				(0.030)
ois2·I (0.277<ois2≤ 0.352)				0.884***	ois2·I (0.889<ois2≤ 0.913)			0.066**
				(0.141)				(0.032)
ois2·I (0.352<ois2≤ 2.652)				0.075***	ois2·I (0.913<ois2≤ 0.981)			-0.051*
				(0.023)				(0.026)
ois2·I (ois2>2.652)				0.023*	ois2·I (ois2>0.981)			-0.023
				(0.013)				(0.019)