Table 10: Variable definitions Variable...

Table 10:

Variable definitions

Variable	Definition	Source
Panel A. Climate change exposure variables
REGEXPO	Firm’s transition risk exposure to climate change, of which the initial bigrams include keywords associated with climate-related regulatory interventions.	Sautner et al. (2023)
PHYEXPO	Firm’s physical risk exposure to climate change, of which the initial bigrams include keywords associated with climate-related physical hazards.	As above
CCEXPO	Firm’s overall exposure to climate change risk, captured by averaging REGEXPO and PHYEXPO.	As above
Panel B. High-leverage variable HLEV	A dummy variable that equals 1 if, in that year, the firm’s long-term debt-to-assets ratio ranks in the top three deciles of the overall sample.	Compustat
Panel C. Outcome variables SALES_ G	Sales growth, equal to percentage growth in sales from the end of the previous year to the end of the current year.	Compustat
EQUITY_ ISSUE	Net equity issuances, equal to equity issuances minus equity repurchases divided by total assets (Compustat items: (SSTK-PRSTKC)/AT).	As above
DEBT_ ISSUE	Net debt issuances, equal to long-term debt issuances minus long-term debt reductions divided by total assets (Compustat items: (DLTIS-DLTR)/AT).	As above
Panel D. Other variables CSR	For each firm-year, we calculate a firm’s overall CSR score by aggregating all seven areas except for corporate governance and obtain CSR using the number of strengths minus the number of concerns.	MSCI ESG STATS
SIZE	Natural logarithm of total assets.	Compustat
PROFIT	Profitability, captured by operating earnings plus depreciation over total assets.	As above
INVESTMENT	Investment, captured by capital expenditures over total assets.	As above
SELLEXP	Selling expenses, the ratio of advertising and selling expenses to total sales.	As above
COGS	Cost of goods sold, the ratio of cost of goods sold to total sales.	As above
COMPETITION	The market share of the four largest firms in a certain industry	As above
PENALTY	The logarithm of 1 plus Penalty by each firm from the given year and the following year. Penalty refers to the dollar amount of penalties.	Violation Tracker
BLUE STATE	Dummy variable that equals 1 if a firm’s headquarters is in a blue state, and 0 otherwise. We classify a state as blue if it voted Democratic in the last presidential election, and red if it voted Republican.	270towin.com
PARIS AGREEMENT	Dummy variable that equals 1 for years after the Paris Agreement was signed (2015), and 0 otherwise.	Compustat
HIGH-TECH, MANUFAC-TURING, and RETAILING	We identify firms with 3-digit SIC codes 283, 357, 366, 367, 382, 384, 481, 482, 489, 737, and 873 as high-tech firms; firms with 2-digit SIC codes ranging from 20 to 39 (excluding firms classified as high-tech firms) as manufacturing firms; and firms with 4-digit SIC codes ranging from 5200 to 5999 as retail firms.	McGurr and De-Vaney (1998); Ho et al. (2005); Kile and Phillips (2009)
Customer Sensitivity	High customer sensitivity industries have the following four-digit SIC codes: [0, 999], [2000, 2399], [2500, 2599], [2700, 2799], [2830, 2869], [3000, 3219], [3420, 3429], 3523, [3600, 3669], [3700, 3719], 3751, [3850, 3879], [3880, 3999], 4813, [4830, 4899], [5000, 5079], [5090, 5099], [5130, 5159], [5220, 5999], [6000, 6999], [7000, 7299], and [7400, 9999]. The remainder are low customer sensitivity industries.	Lev et al. (2010)
Product Differentiation	We rely on product similarity created by the Hoberg—Phillips Data Library. We define a firm as having low (high) product differentiation if the product similarity score is above (below or equal to) the median two years before the base year.	Hoberg—Phillips Data Library
Product Market Competition	Low (High) product market competition refers to the Herfindahl index of firm sales above (below) the overall sample median two years before the base year.	Compustat
Labor Intensity	We classify an industry as high (low) labor intensity if the wage-to-capital ratio is higher (lower) than the median for all industries two years before the base year.	As above

Variable	Definition	Source
Panel A. Climate change exposure variables
REGEXPO	Firm’s transition risk exposure to climate change, of which the initial bigrams include keywords associated with climate-related regulatory interventions.	Sautner et al. (2023)
PHYEXPO	Firm’s physical risk exposure to climate change, of which the initial bigrams include keywords associated with climate-related physical hazards.	As above
CCEXPO	Firm’s overall exposure to climate change risk, captured by averaging REGEXPO and PHYEXPO.	As above
Panel B. High-leverage variable HLEV	A dummy variable that equals 1 if, in that year, the firm’s long-term debt-to-assets ratio ranks in the top three deciles of the overall sample.	Compustat
Panel C. Outcome variables SALES_ G	Sales growth, equal to percentage growth in sales from the end of the previous year to the end of the current year.	Compustat
EQUITY_ ISSUE	Net equity issuances, equal to equity issuances minus equity repurchases divided by total assets (Compustat items: (SSTK-PRSTKC)/AT).	As above
DEBT_ ISSUE	Net debt issuances, equal to long-term debt issuances minus long-term debt reductions divided by total assets (Compustat items: (DLTIS-DLTR)/AT).	As above
Panel D. Other variables CSR	For each firm-year, we calculate a firm’s overall CSR score by aggregating all seven areas except for corporate governance and obtain CSR using the number of strengths minus the number of concerns.	MSCI ESG STATS
SIZE	Natural logarithm of total assets.	Compustat
PROFIT	Profitability, captured by operating earnings plus depreciation over total assets.	As above
INVESTMENT	Investment, captured by capital expenditures over total assets.	As above
SELLEXP	Selling expenses, the ratio of advertising and selling expenses to total sales.	As above
COGS	Cost of goods sold, the ratio of cost of goods sold to total sales.	As above
COMPETITION	The market share of the four largest firms in a certain industry	As above
PENALTY	The logarithm of 1 plus Penalty by each firm from the given year and the following year. Penalty refers to the dollar amount of penalties.	Violation Tracker
BLUE STATE	Dummy variable that equals 1 if a firm’s headquarters is in a blue state, and 0 otherwise. We classify a state as blue if it voted Democratic in the last presidential election, and red if it voted Republican.	270towin.com
PARIS AGREEMENT	Dummy variable that equals 1 for years after the Paris Agreement was signed (2015), and 0 otherwise.	Compustat
HIGH-TECH, MANUFAC-TURING, and RETAILING	We identify firms with 3-digit SIC codes 283, 357, 366, 367, 382, 384, 481, 482, 489, 737, and 873 as high-tech firms; firms with 2-digit SIC codes ranging from 20 to 39 (excluding firms classified as high-tech firms) as manufacturing firms; and firms with 4-digit SIC codes ranging from 5200 to 5999 as retail firms.	McGurr and De-Vaney (1998); Ho et al. (2005); Kile and Phillips (2009)
Customer Sensitivity	High customer sensitivity industries have the following four-digit SIC codes: [0, 999], [2000, 2399], [2500, 2599], [2700, 2799], [2830, 2869], [3000, 3219], [3420, 3429], 3523, [3600, 3669], [3700, 3719], 3751, [3850, 3879], [3880, 3999], 4813, [4830, 4899], [5000, 5079], [5090, 5099], [5130, 5159], [5220, 5999], [6000, 6999], [7000, 7299], and [7400, 9999]. The remainder are low customer sensitivity industries.	Lev et al. (2010)
Product Differentiation	We rely on product similarity created by the Hoberg—Phillips Data Library. We define a firm as having low (high) product differentiation if the product similarity score is above (below or equal to) the median two years before the base year.	Hoberg—Phillips Data Library
Product Market Competition	Low (High) product market competition refers to the Herfindahl index of firm sales above (below) the overall sample median two years before the base year.	Compustat
Labor Intensity	We classify an industry as high (low) labor intensity if the wage-to-capital ratio is higher (lower) than the median for all industries two years before the base year.	As above

[ViewLarge]

Sharing Unavailable