Replication data for: Leviathan Inc. and corporate environmental engagement

This package contains publicly sharable data, proprietary data (for internal replication only) and Stata codes for the paper Hsu, Liang, Matos (2021) "Leviathan Inc. and Corporate Environmental Engagement", accepted for publication at Management Science.Data and Code Availability:* We have prepared the following documents:- "Codes_HsuLiangMatos_MS.do" contains the Stata codes that produce all our results in the paper;- "data_for_public_MS.dta" contains the dataset (in Stata format) that can be shared to the public as these data are not protected by Confidentiality Agreement with the data providers/vendors;- "data_proprietary_MS.dta" contains the entire dataset (in Stata format) for our main tests, with some of the data not to be shared to the public as they are protected by Confidentiality Agreement with the data providers/vendors. They are for the purpose of internal replication only;- "msci_esg.dta" contains the dataset (in Stata format) for our robustness test in Table 11, with most of the data not to be shared to the public as they are protected by Confidentiality Agreement with the data providers/vendors;- "sustainalytics.dta" contains the dataset (in Stata format) for our robustness test in Table 11, with most of the data not to be shared to the public as they are protected by Confidentiality Agreement with the data providers/vendors;- "forbes.dta" contains the dataset of Forbes Top 30 Global Companies in 2010 that is used to create Table 1 in the paper;- "Figure1_data.xlsx" contains the processed data that is used to create Figure 1 in the paper;- "Figures3&4_data.xlsx" contains the processed data that is used to create Figure 3 and Figure 4 in the paper;- "Log_file.smcl" contains the log file for running all the codes in "Codes_HsuLiangMatos_MS.do".**** Data for Public **** * The following variables are collected from sources that are publicly available; thus they can be shared to the public (included in "data_for_public_MS.dta"): - GUO_state3 : measure for our main dependent variable "State_own" - a binary variable indicating the ultimate owner of the firm is a state (Source: Orbis + authors' mannual correction based on company annual reports & websites)- government_held3a : percentage of free-floating shares held by the government as a blockholder (owning > 5% stake) (Source: Datastream)- lngdppc : The logarithm of GDP per capita of the country (Source: World Bank)- gov1_01a : A binary variable indicating that the largest government party orientation wrt economic policy changed from center/right to left (Source: Database of Political Institutions)- gov1_10a : A binary variable indicating that the largest government party orientation wrt economic policy changed from left to center/right (Source: Database of Political Institutions)- energy_dependence : International Index of Energy Security Risk by Global Energy Institute (Source: Global Energy Institution; https://www.globalenergyinstitute.org/energy-security-risk-index)- energy_dependence_hi : A binary variable indicating a country's energy_dependence index value is greater than the sample median- mktcap_gdp : market capitalisation to GDP ratio (Source: World bank)- mktcap_gdp_hi : A binary variable indicating a country's mktcap_gdp ratio is greater than the sample median- E_regulation_hi : A binary variable indicating a country's environmental regulation index is greater than the sample median (Source: Carrots & Sticks and Schiller (2018) Working Paper "Global Supply-Chain Networks and Corporate Social Responsibility")- nuclear : A binary variable indicating a firm is a nuclear company (Source： World Nuclear Association)- vae : Voice & Accountability (Estimate) Index from World Governance Indicators (Source: World Bank - World Governance Indicators)- pve : Political Stability and Absence of Violence/Terrorism (Estimate) Index from World Governance Indicators (Source: World Bank - World Governance Indicators)- gee : Government Effectiveness (Estimate) Index from World Governance Indicators (Source: World Bank - World Governance Indicators)- rqe : Regulatory Quality (Estimate) Index from World Governance Indicators (Source: World Bank - World Governance Indicators)- rle : Rule of Law (Estimate) Index from World Governance Indicators (Source: World Bank - World Governance Indicators)- cce : Control of Corruption (Estimate) Index from World Governance Indicators (Source: World Bank - World Governance Indicators)**** Proprietary Data (for replication only) ***** The following variables are collected from *proprietary* data sources (commercial databases subscribed by the authors' institutions); they are included in "data_proprietary_MS.dta" (NOT to be shared to the public);* Here, we provide the simple definitions of them. Interested readers are referred to the original data sources to collect them:- ENVSCORE : Overall Environmental Pillar Score from Thomson Reuters ASSET4- ENER : Emission reduction score (a component score of ENVSCORE) from Thomson Reuters ASSET4- ENPI : Production innovation score (a component score of ENVSCORE) from Thomson Reuters ASSET4- ENRR : Resource Reduction score (a component score of ENVSCORE) from Thomson Reuters ASSET4- CGVSCORE : Overall Corporate Governance Pillar Score from Thomson Reuters ASSET4- SOCSCORE : Overall Social Pillar Score from Thomson Reuters ASSET4- io : Total institutional ownership of a company (data from FactSet)- mktcap : total market capitalization of a firm's outstanding equity (data from Datastream/Worldscope)- leverage : debt to assets ratio of a firm (data from Datastream/Worldscope)- MTBV2 : Market-to-book ratio of a firm's total equity (item8001/item3501) (data from Datastream/Worldscope)- roa : return on assets (net income/total assets) (data from Datastream/Worldscope)- hhi48_ctry : The Herfindahl-Hirschman Index (HHI) based on Fama-French 48 industries classification for each country- ln_total_co2_a : The logarithm of firm-level total co2 emission to total assets ratio; data on total co2 emission is obtained from ASSET4- waste_sic2_med2 : Industry-level emission intensity - we use ASSET4_SIC and TRI database to assign SIC2 industries into high and low emission- mean_ab_temp_abs_hi : A binary indicator for whether a city's absolute value of abnormal temperature (annual average) is above sample median- mean_pdsi_hi : A binary indicator of whether a country's Palmer Drought Severity Index value (annual average) is above sample median (annual average)- foreign_assets : Percentage of a firm's foreign assets over its total assets (data from Datastream/Worldscope)- foreign_sales : Percentage of a firm's foreign sales over its total sales (data from Datastream/Worldscope)- foreign_income_growth : Foreign income growth (data from Datastream/Worldscope)- retirement_age1 : A binary variable indicating whether the CEO is close to retirement age (at least 60 years old) (data from NRG Metrics and BoardEx)- Tenure : CEO's tenure in the company (data from NRG Metrics & BoardEx)- PoliticalConnection : A binary indicator for whether the CEO has any form of political connection (data from NRG Metrics and BoardEx)- government_held : The percentage of total shares in issue of holdings of 5% or more held by a government or government institution (data from Datastream/Worldscope)- foreign_holdings : The percentage of total shares in issue of holdings of 5% or more held by a foreign institution (data from Datastream/Worldscope)- cross_holdings : The percentage of total shares in issue of holdings of 5% or more held by one company in another (data from Datastream/Worldscope)- pension_fund_held : The percentage of total shares in issue of holdings of 5% or more held by pension funds or endowment funds (data from Datastream/Worldscope)- investment_co_held : The percentage of total shares in issue of holdings of 5% or more held as long term strategic holdings by investment banks (data from Datastream/Worldscope)- employee_held : The percentage of total shares in issue of 5% or more held by employees (data from Datastream/Worldscope)- strategic_holdings : The percentage of total shares in issue of 5% or more held strategically and not available to ordinary investors (data from Datastream/Worldscope)- io_dom : A firm's domestic institutional ownership (data from Factset Lionshare)- io_for : A firm's foreign institutional ownership (data from Factset Lionshare)- foreign_GUO : A binary variable indicating a firm's ultimate owner is a foreign government or government entity (data from Orbis + authors' self calculation)- swf : A binary variable indicating a firm has sovereign wealth fund ownership (data from Factset and SWFI)- trbc_code: Thomson Reuters (Refinitive) Business Classification code (data froM Datastream)* The following commands may need to be installed on Stata to run the codes and export the results as in "Codes_HsuLiangMatos_MS.do":- asdoc- winsor- carryforward- reghdfe- outreg2

本数据包包含可公开分享的数据、专有数据（仅限内部复制使用）以及论文《Hsu, Liang, Matos（2021）中的 Stata 代码：“Leviathan Inc. and Corporate Environmental Engagement”，该论文已被《管理科学》杂志接受发表。数据与代码的可访问性如下：* 我们已准备以下文件：- “Codes_HsuLiangMatos_MS.do” 包含生成论文中所有结果的 Stata 代码；- “data_for_public_MS.dta” 包含数据集（Stata 格式），由于这些数据未受到与数据提供者/供应商的保密协议保护，因此可以公开分享；- “data_proprietary_MS.dta” 包含我们主要测试的全部数据集（Stata 格式），其中部分数据因受到与数据提供者/供应商的保密协议保护而不宜公开分享，仅供内部复制使用；- “msci_esg.dta” 包含我们第 11 表稳健性测试的数据集（Stata 格式），其中大部分数据因受到与数据提供者/供应商的保密协议保护而不宜公开分享；- “sustainalytics.dta” 包含我们第 11 表稳健性测试的数据集（Stata 格式），其中大部分数据因受到与数据提供者/供应商的保密协议保护而不宜公开分享；- “forbes.dta” 包含用于创建论文中第 1 表的 2010 年《福布斯》全球 30 强公司数据集；- “Figure1_data.xlsx” 包含用于创建论文中第 1 图的数据；- “Figures3&4_data.xlsx” 包含用于创建论文中第 3 和第 4 图的数据；- “Log_file.smcl” 包含运行“Codes_HsuLiangMatos_MS.do”中所有代码的日志文件。**** 公开数据 **** * 以下变量来自公开可获得的数据源，因此可以公开分享（包含在“data_for_public_MS.dta”中）：- GUO_state3：衡量我们主要因变量“State_own”的指标，为一个二元变量，表示公司的最终所有者是政府（来源：Orbis + 作者根据公司年报和网站进行的手册修正）- government_held3a：政府作为大股东持有的非流通股份比例（持有超过 5% 的股份）（来源：Datastream）- lngdppc：人均 GDP 的对数（来源：世界银行）- gov1_01a：一个二元变量，表示最大的政府政党在经济政策方面的最大政府政党方向从中心/右转向左（来源：政治机构数据库）- gov1_10a：一个二元变量，表示最大的政府政党在经济政策方面的最大政府政党方向从左转向中心/右（来源：政治机构数据库）- energy_dependence：全球能源研究所发布的国际能源安全风险指数（来源：全球能源研究所；https://www.globalenergyinstitute.org/energy-security-risk-index）- energy_dependence_hi：一个二元变量，表示一个国家的 energy_dependence 指数值大于样本中位数- mktcap_gdp：市值与 GDP 比率（来源：世界银行）- mktcap_gdp_hi：一个二元变量，表示一个国家的 mktcap_gdp 比率大于样本中位数- E_regulation_hi：一个二元变量，表示一个国家的环境规制指数大于样本中位数（来源：Carrots & Sticks 和 Schiller（2018）工作论文“全球供应链网络与企业社会责任”）- nuclear：一个二元变量，表示一家公司是核能公司（来源：世界核能协会）- vae：来自世界治理指标的语音与问责制（估计）指数（来源：世界银行 - 世界治理指标）- pve：来自世界治理指标的政治稳定性和无暴力/恐怖主义（估计）指数（来源：世界银行 - 世界治理指标）- gee：来自世界治理指标的政府有效性（估计）指数（来源：世界银行 - 世界治理指标）- rqe：来自世界治理指标的规制质量（估计）指数（来源：世界银行 - 世界治理指标）- rle：来自世界治理指标的法治（估计）指数（来源：世界银行 - 世界治理指标）- cce：来自世界治理指标的腐败控制（估计）指数（来源：世界银行 - 世界治理指标）**** 专有数据（仅限复制使用） ***** 以下变量来自 *专有* 数据源（作者机构订阅的商业数据库）；它们包含在“data_proprietary_MS.dta”中（不得公开分享）；* 在此处，我们提供了它们的基本定义。感兴趣的读者可参考原始数据源以获取它们：- ENVSCORE：来自汤姆森路透 ASSET4 的整体环境支柱得分- ENER：ENVSCORE 的一个组成部分，即排放减少得分- ENPI：ENVSCORE 的一个组成部分，即生产创新得分- ENRR：ENVSCORE 的一个组成部分，即资源减少得分- CGVSCORE：来自汤姆森路透 ASSET4 的整体公司治理支柱得分- SOCSCORE：来自汤姆森路透 ASSET4 的整体社会支柱得分- io：公司的总机构所有权（FactSet 数据）- mktcap：公司未偿还股票的总市值（Datastream/Worldscope 数据）- leverage：公司的债务与资产比率（Datastream/Worldscope 数据）- MTBV2：公司总股权的市场对账面比率（Datastream/Worldscope 数据）- roa：资产回报率（净收入/总资产）（Datastream/Worldscope 数据）- hhi48_ctry：基于 Fama-French 48 个行业分类的每个国家的 Herfindahl-Hirschman 指数（HHI）- ln_total_co2_a：公司层面总二氧化碳排放与总资产比率的对数；总二氧化碳排放数据来自 ASSET4- waste_sic2_med2：行业层面的排放强度 - 我们使用 ASSET4_SIC 和 TRI 数据库将 SIC2 行业分配到高排放和低排放- mean_ab_temp_abs_hi：一个二元指标，表示一个城市的异常温度（年度平均值）的绝对值是否高于样本中位数- mean_pdsi_hi：一个二元指标，表示一个国家的 Palmer Drought Severity Index 值（年度平均值）是否高于样本中位数（年度平均值）- foreign_assets：公司外国资产占总资产的比例（Datastream/Worldscope 数据）- foreign_sales：公司外国销售额占总销售额的比例（Datastream/Worldscope 数据）- foreign_income_growth：外国收入增长（Datastream/Worldscope 数据）- retirement_age1：一个二元变量，表示 CEO 是否接近退休年龄（至少 60 岁）（NRG Metrics 和 BoardEx 数据）- Tenure：CEO 在公司的任期（NRG Metrics & BoardEx 数据）- PoliticalConnection：一个二元指标，表示 CEO 是否有任何形式的政界联系（NRG Metrics 和 BoardEx 数据）- government_held：持有 5% 或更多股份的政府或政府机构持有的发行在外的总股份比例（Datastream/Worldscope 数据）- foreign_holdings：持有 5% 或更多股份的外国机构持有的发行在外的总股份比例（Datastream/Worldscope 数据）- cross_holdings：一个公司在另一个公司中持有 5% 或更多股份的发行在外的总股份比例（Datastream/Worldscope 数据）- pension_fund_held：持有 5% 或更多股份的养老基金或捐赠基金持有的发行在外的总股份比例（Datastream/Worldscope 数据）- investment_co_held：作为长期战略持股持有的发行在外的总股份比例的 5% 或更多股份（Datastream/Worldscope 数据）- employee_held：持有发行在外的 5% 或更多股份的员工的百分比（Datastream/Worldscope 数据）- strategic_holdings：持有发行在外的 5% 或更多股份的战略持股，不可供普通投资者获取（Datastream/Worldscope 数据）- io_dom：公司的国内机构所有权（Factset Lionshare 数据）- io_for：公司的外国机构所有权（Factset Lionshare 数据）- foreign_GUO：一个二元变量，表示公司的最终所有者是外国政府或政府实体（Orbis + 作者自行计算）- swf：一个二元变量，表示公司拥有主权财富基金所有权（Factset 和 SWFI 数据）- trbc_code：汤姆森路透（Refinitive）商业分类代码（Datastream 数据）* 可能需要在 Stata 上安装以下命令以运行代码并将结果导出为“Codes_HsuLiangMatos_MS.do”中的格式：- asdoc- winsor- carryforward- reghdfe- outreg2