## Appendix 6: Sustainability-linked Loan Borrowers vs Non-borrowers by Industry ## Source details **Canonical URL:** [Appendix 6: Sustainability-linked Loan Borrowers vs Non-borrowers by Industry](https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021194-print-pdf.pdf) ## Other formats - [Markdown version](/-/media/files/publications/wp/2021/english/wpiea2021194-print-pdf.pdf.md) - [Structured JSON version](/-/media/files/publications/wp/2021/english/wpiea2021194-print-pdf.pdf.json) --- ### I. Introduction — purpose and scope - Green debt markets are rapidly growing alongside investor interest in climate change issues. - Paper objective: overview green debt products and market developments for corporate issuers and present data on CO2 emissions of green debt issuers. - Key analytic expectation: issuers of green debt would outperform peers on CO2 emission reductions if green debt is achieving its environmental objectives; CO2 emissions are used as a firm-level measure despite being a crude proxy that does not imply causality. - Terminology: "green debt" denotes all sustainable debt with environmental objectives, noting most sustainable debt outstanding targets climate change. ### II. Definitions and classification of sustainable debt instruments - Two main categories of sustainable debt instruments: - Activity-based instruments: issuer must use proceeds for specified sustainable projects; issuer’s overall sustainability profile not required. Examples: green bonds, social bonds, sustainability bonds, green loans. - Issuer-based instruments: include entity-wide sustainability performance targets; proceeds may be used for general purposes; interest rates typically vary depending on achievement of targets. Examples: sustainability-linked bonds and sustainability-linked loans. - Cumulative issuance up to 2020 ($bn): - Green bond — Activity-based — Environmental projects — 1,139 - Social bond — Activity-based — Social projects — 192 - Sustainability bond — Activity-based — Environmental & social projects — 143 - Sustainability-linked bond — Issuer-based — Institutional ESG targets — 16 - Green loan — Activity-based — Environmental projects — 501 - Sustainability-linked loan — Issuer-based — Institutional ESG targets — 311 - Source: Adapted from BloombergNEF; IMF staff. Data as of December 31, 2020. ### III. Governance, standards, and external review - Sustainable debt relies primarily on issuer self-labeling supplemented by voluntary principles, guidelines, and standards. - Key private governance players: - ICMA: principle frameworks for green bonds, social bonds, sustainability bonds, and sustainability-linked bonds. - Asia Pacific Loan Market Association, Loan Market Association, Loan Syndications and Trading Association: guidance for green loans and sustainability-linked loans. - Climate Bonds Initiative (CBI): standards and certification process for green bonds. - External reviews: - External reviews common for green bonds; uncommon for sustainability-linked bonds and loans; green loans typically have no external reviews. - Empirical notes: certification associated with lower yields/greater investor interest and larger post-issuance CO2 reductions for certified issuers (cited studies). - Regulatory developments: - Regulators increasingly setting standards (examples cited: China, India, Japan, ongoing EU work). - Regulators also influence markets via taxonomies and support schemes (examples cited in source). ### IV. Risks: greenwashing and reporting challenges - Voluntary standards and self-labeling raise risk of greenwashing ex ante and ex post (failures in reporting, monitoring, impact assessment). - Proliferation of frameworks and standards can create complexity, potentially hindering market development and investor assessment. ### V. Credit risk and structural decompositions - Credit risk of sustainable/green debt equals conventional debt by the same issuer. - Conceptual decompositions: - Activity-based instruments: conventional debt + promise to use proceeds for earmarked purposes. - Issuer-based instruments: conventional debt + contractual interest-rate adjustments tied to sustainability targets (conceptually a conventional bond plus a certificate with penalty/bonus contingent on target achievement). - Sovereign structural innovations to preserve liquidity and signal greenness: - Danish proposal: decompose green bond into conventional bond + tradable green certificate. - German twin bond concept: issue conventional and green bond twins with same coupon and maturity allowing swaps to preserve liquidity while creating a green curve. ### VI. Motivations for corporates to issue green debt - Issuance imposes additional costs relative to conventional debt (certification fees, reporting, restrictions). - Potential issuer rationales: - Pricing advantage: mixed empirical evidence; literature inconclusive with findings of premiums, discounts, and no premium depending on sample and methodology. - Studies cited with different findings (examples in source). - Signaling credible commitment to environment: evidence shows positive stock market reactions, increased investor holdings, and post-issuance environmental performance improvements, especially for certified and first-time issuers. - Greenwashing motive considered but some empirical evidence suggests post-issuance environmental improvements, inconsistent with pure greenwashing. - Other motives from issuer interviews: investor demand, broaden investor base, anticipate future capital-access importance of green credentials, and internal mainstreaming of sustainability; underwriting banks often initiate issuance. ### VII. Motivations for investors to buy green debt - Financial return considerations appear secondary; evidence does not show a consistent green premium. - Investor motives include: - Signaling green credentials to end-investors and internally (asset managers). - Potential risk reduction through issuers’ better sustainability performance (debated). - Product marketing advantages for intermediaries (justify higher fees). - Activist and client pressure to align portfolios with green/climate targets. - Engagement opportunities with issuers and end-investor demand. ### VIII. Environmental impact channels - Activity-based instruments: - Direct channel (additionality): funding environmentally valuable projects that would not be financed without green debt. Two necessary conditions: (1) funded projects must be environmentally valuable; (2) financing leads to projects that would not otherwise be financed. Pure cost-of-capital additionality deemed unlikely because green debt yields are generally similar to conventional debt. - Indirect channels: internal awareness and capacity building, investor/stakeholder oversight and engagement, and fostering an investable climate/green ecosystem (taxonomies, standards, certification). - Issuer-based instruments (sustainability-linked): - Direct channel: interest-rate link to sustainability targets provides incentive; effectiveness depends on ambition and clarity of targets and additionality concerns. - Indirect channels: incentivize integration of sustainability objectives across the issuer’s agenda; issuer-level commitments may foster organizational changes. - Example from source: a sustainability-linked bond issued by ENI includes a 25 bps interest rate step up if carbon footprint and renewable capacity targets are not reached. ### IX. Literature highlights and empirical findings (selected) - Literature focused mainly on green bonds; mixed findings on pricing and environmental impact. - Key empirical findings summarized: - Flammer (2021): green bond issuers improve environmental performance post issuance; certified issuances receive stronger investor interest and greater CO2 reductions. - Ehlers et al. (2020): find no clear evidence of lower or decreasing carbon intensity for green bond issuers; recommend complementing green bonds with issuer-level emissions ratings. - Tuhkanen and Vulturius (2020): many large European green bond issuers do not link green bond frameworks to climate targets; highlight shortcomings in post-issuance reporting raising greenwashing risks. - Open questions: measurement of firm-level emission impacts, additionality, role of external reviews/certification for sustainability-linked instruments, and investor motivations. ### III. Sustainable and green debt market developments (key statistics) - Sustainable debt issuance reached $730bn in 2020. - Total global debt at $281tn (Institute of International Finance). - Activity-based instruments accounted for more than 80 percent of sustainable debt issued in 2020. - Issuer-based instruments reached a share of about 20 percent of sustainable debt in 2020 from close to zero in 2016. - Regional shares in 2020: - Europe accounted for more than half of sustainable debt issuance. - Western Hemisphere at 20 percent. - Asia at 15 percent. - Green bonds: - About $300bn raised in 2020. - European issuers accounted for 55% of issuance. - EUR and USD combined account for 80 percent of issuance. - Corporates account for slightly more than half of all green bonds issued. - National governments account for about 22 percent; asset-backed securities (ABS) account for 10 percent; supranational issuers 9 percent; US municipal debt 5 percent. - Green loans: - Volume in 2020 about $80bn, down from $93bn in 2019. - Almost exclusively used by corporates, mainly energy and utilities companies. - Sustainability-linked loans and bonds: - About $120bn of sustainability-linked loans in 2020, from $140bn in 2019. - Sustainability-linked bonds: total issuance in 2020 was $10.6bn. - ICMA published formal guidelines for sustainability-linked loans in June 2020 and sustainability-linked bond principles in June 2020. - Tanzania became the first sovereign borrower with a sustainability linkage in 2020. - Enel issued the first corporate sustainability-linked bond in 2019. ### IV. Greenness of issuers — data, measures, and findings - Greenness measurement: - Measured by CO2 emissions scaled by revenue and assets. - Emission reporting scopes: scope 1 (direct), scope 2 (indirect from purchased energy), scope 3 (other indirect value-chain emissions). - Presented data for scope 1 and 2 emissions; results robust to inclusion of scope 3. - Dataset and coverage: - Includes all firms in the S&P Global 1200 index. - Among S&P Global 1200 firms, 76% report at least scope 1 and 2 emission data. - CO2 emissions from Bloomberg and Thomson Reuters; revenues/assets from Bloomberg; green debt data from BNEF; environmental scores from Refinitiv via Thomson Reuters. - Observations: 148 unique firm-level observations for green bond issuers, 49 observations for green loan borrowers, and 70 observations for sustainability-linked loan borrowers. - Summary statistics (selected): - Green bond issuers are larger than average S&P Global 1200 firms by revenues and especially by assets. - Over half of green bond issuers are financials or utilities. - Green loan borrowers tend to be larger than average firms but less asset-heavy than green bond issuers; dominated by utilities (45%), followed by financials (20%). - Sustainability-linked loan borrowers: similar in revenues to average index firm and smaller by assets; industry structure dominated by utilities (21%), industrials (20%), and materials (17%). - Empirical findings by instrument: - Green Bonds: - Green bond issuers have lower CO2 emissions relative to revenue and assets than firms that do not issue green bonds. - Green bond issuers are more likely to report CO2 emissions. - Results robust to emission scope and persist within industries with at least 10 observations (financials, utilities, industrials, consumer discretionary). - Green bond issuers tend to have higher environmental rating scores than non-issuers. - Green bond issuers that engage external reviewers for assurance tend to have lower CO2 emission intensities than those without external assurance. - Evolution around first issuance: emission intensities reduce notably in the second year after first issuing a green bond; significant reduction in the year of issuance relative to the previous year. - Green Loans: - Green loan borrowers tend to have higher CO2 emissions relative to revenue and assets than firms that do not borrow with green loans; driven by industry composition (utilities overrepresented). - Within industries, no clear difference in emission intensities between green loan borrowers and other firms. - Evolution around loan origination: improvement from two years before to two years after origination, primarily due to a drop in emission intensities in the second year following the loan; overall smaller reduction over the 5-year window compared with green bond issuers. - Possible explanations: classification and standards differences, limited external assurance, less ambitious green conditions and monitoring. - Sustainability-linked Loans: - Borrowers tend to have somewhat higher CO2 emission intensities in aggregate, driven by industry composition. - For utilities (21% of sample), sustainability-linked borrowers have somewhat lower emission intensities than other firms. - Evolution around loan: big improvement in CO2/revenue and CO2/assets in the year of the loan and some improvement in the following year; improvement larger than for green loans and similar magnitude as for green bonds. - Tentative interpretation: sustainability-linked loans may incentivize emissions reductions somewhat better than green loans; alternatively, greener firms may be more interested in sustainability-linked loans. - Data limitations: small sample sizes due to novelty; only three years of evolution possible given first loan in 2017; insufficient observations to analyze sustainability-linked bonds (exist since 2019). ### V. Conclusion — synthesis and research needs - Market dynamics: - Green debt markets are rapidly evolving with growth in issuance, security design development, and refinement of governance, standards, and guidelines. - Green bonds remain the largest sustainable debt category; issuer-based instruments linking interest rates to firm-level sustainability targets have emerged. - Role in policy mix: - Many policymakers and investors view green debt as an important component to achieve the transition to a low carbon economy and ensure pricing of climate risks. - Empirical synthesis: - Documented CO2 emission intensity of green debt issuers relative to other firms and over time: - Green bond issuers have lower emission intensities than other firms. - Green loan and sustainability-linked loan borrowers are in line with other firms after controlling for industry composition. - Green bond, green loan, and sustainability-linked loan borrowers lower their emission intensity over time at a faster rate than other firms. - Likely interpretation: issuers pursue green debt to signal green credentials; engagement with investors, organizational learning, and mainstreaming green considerations also potentially contribute. - Direct financial benefits through lower debt cost are unlikely to be a major factor given lack of conclusive evidence of substantive pricing differences between green and conventional debt. - Research needs: - More research on environmental impact and its channels, motivations of market participants, and design of green debt to incentivize environmental objectives. - Comparative analysis of activity-based vs issuer-based instruments and of market-based versus bank-based debt financing as more data become available. *Source: IMF staff (content unit: Appendix 6: Sustainability-linked Loan Borrowers vs Non-borrowers by Industry, from wpiea2021194-print-pdf).* ### References.............................................................................................................. ### References ### References section - References....................................................................................................................... 23 ### Appendices overview - Appendix 1: Green Bond Issu e rs v s Non-issuers by Scope of Emissions ............................ 25 - Appendix 2: Green Bond Issuers vs Non-issuers by Industry ............................................. 26 - Appendix 3: Green Loan Borrowers vs Non-borrowers by Scope of Emissions .................. 27 - Appendix 4: Green Loan Borrowers vs Non-borrowers by Industry ................................... 28 - Appendix 5: Sustainability-linked Loan Borrowers vs Non-borrowers by Scope of Emissions ....................................................................................................................................... 29 *wpiea2021194-print-pdf - References..............................................................................................................* ### Appendix 6: Sustainability-linked Loan Borrowers vs Non-borrowers by Industry ............. 30 ### Appendix 6: Sustainability-linked Loan Borrowers vs Non-borrowers by Industry ### I. Introduction — purpose and scope - Green debt markets are rapidly growing alongside investor interest in climate change issues. - Paper objective: overview green debt products and market developments for corporate issuers and present data on CO2 emissions of green debt issuers. - Key analytic expectation: issuers of green debt would outperform peers on CO2 emission reductions if green debt is achieving its environmental objectives; CO2 emissions are used as a firm-level measure despite being a crude proxy that does not imply causality. - Terminology: "green debt" used to denote all sustainable debt with environmental objectives, noting most sustainable debt outstanding targets climate change. ### II. Definitions and classification of sustainable debt instruments - Two main categories of sustainable debt instruments: - Activity-based instruments: issuer must use proceeds for specified sustainable projects; issuer’s overall sustainability profile not required. Examples: green bonds, social bonds, sustainability bonds, green loans. - Issuer-based instruments: include entity-wide sustainability performance targets; proceeds may be used for general purposes; interest rates typically vary depending on achievement of targets. Examples: sustainability-linked bonds and sustainability-linked loans. - Cumulative issuance up to 2020 ($bn): - Green bond — Activity-based — Environmental projects — 1,139 - Social bond — Activity-based — Social projects — 192 - Sustainability bond — Activity-based — Environmental & social projects — 143 - Sustainability-linked bond — Issuer-based — Institutional ESG targets — 16 - Green loan — Activity-based — Environmental projects — 501 - Sustainability-linked loan — Issuer-based — Institutional ESG targets — 311 - Source noted: Adapted from BloombergNEF; IMF staff. Data as of December 31, 2020. ### III. Governance, standards, and external review - Sustainable debt relies primarily on issuer self-labeling supplemented by voluntary principles, guidelines, and standards. - Key private governance players: - ICMA provides principle frameworks for green bonds, social bonds, sustainability bonds, and sustainability-linked bonds. - Asia Pacific Loan Market Association, Loan Market Association, Loan Syndications and Trading Association: guidance for green loans and sustainability-linked loans. - Climate Bonds Initiative (CBI): standards and certification process for green bonds. - External reviews: - Various pre- and post-issuance external review services exist, typically for fees. - External reviews are common for green bonds but uncommon for sustainability-linked bonds and loans; green loans typically have no external reviews. - Empirical notes: Baker et al. (2018) and Flammer (2021) find certification associated with lower yields/greater investor interest and larger post-issuance CO2 reductions for certified issuers. - Regulatory developments: - Regulators increasingly setting standards (examples cited: China, India, Japan, ongoing EU work). - Regulators also influence markets via taxonomies and support schemes (examples: Monetary Authority of Singapore grants; Japan Ministry of the Environment subsidies). ### IV. Risks: greenwashing and reporting challenges - Voluntary standards and self-labeling raise risk of greenwashing both ex ante and ex post (failures in reporting, monitoring, impact assessment). - Proliferation of frameworks and standards can create complexity, potentially hindering market development and investor assessment. ### V. Credit risk and structural decompositions - Credit risk of sustainable/green debt equals conventional debt by the same issuer. - Conceptual decompositions: - Activity-based instruments: conventional debt + promise to use proceeds for earmarked purposes. - Issuer-based instruments: conventional debt + contractual interest-rate adjustments tied to sustainability targets (can be conceptually decomposed into a conventional bond plus a certificate with penalty/bonus contingent on target achievement). - Sovereign examples of structural innovations to preserve liquidity and signal greenness: - Danish proposal: decompose green bond into conventional bond + tradable green certificate. - German twin bond concept: issue conventional and green bond twins with same coupon and maturity allowing swaps to preserve liquidity while creating a green curve. ### VI. Motivations for corporates to issue green debt - Issuance imposes additional costs relative to conventional debt (certification fees, reporting, restrictions). - Potential issuer rationales: - Pricing advantage: mixed empirical evidence on green premium; literature inconclusive with findings including premiums, discounts, and no premium depending on sample and methodology. - Studies finding premium: Baker et al. (2018); Zerbib (2019); Kapraun and Scheins (2019). - Studies finding discount or no premium: Karpf and Mandel (2017); Larcker and Watts (2019); IMF (2019); Flammer (2021). - Maltais and Nykvist (2020): cost of capital considerations are secondary in issuer decisions. - Signaling credible commitment to environment: empirical evidence (Flammer 2021; Tang and Zhang 2020) shows positive stock market reactions, increased investor holdings, and post-issuance environmental performance improvements, especially for certified and first-time issuers. - Greenwashing motive considered but empirical evidence (Flammer 2021) suggests post-issuance environmental improvements, which is inconsistent with a pure greenwashing explanation. - Other motives from issuer interviews (Maltais and Nykvist 2020): investor demand, broaden investor base, anticipate future capital-access importance of green credentials, and internal mainstreaming of sustainability; underwriting banks often initiate issuance. ### VII. Motivations for investors to buy green debt - Financial return considerations appear secondary; evidence does not show a consistent green premium. - Investor motives include: - Signaling green credentials to end-investors and internally (asset managers). - Potential (but debated) risk reduction through issuers’ better sustainability performance. - Product marketing advantages for intermediaries (justify higher fees). - Activist and client pressure to align portfolios with green/climate targets. - Engagement opportunities with issuers and end-investor demand. - Overall: limited empirical study on investor motives; available evidence shows signaling, engagement, and organizational internalization are prominent drivers. ### VIII. Environmental impact channels (summary of channels discussed) - Activity-based instruments: - Direct channel: funding environmentally valuable projects that would otherwise not be undertaken (additionality). Two necessary conditions: (1) funded projects must be environmentally valuable (avoid greenwashing); (2) financing leads to projects that would not otherwise be financed. Pure cost-of-capital additionality deemed unlikely because green debt yields are generally similar to conventional debt. - Indirect channels: internal awareness and capacity building, investor/stakeholder oversight and engagement, and fostering an investable climate/green ecosystem (taxonomies, standards, certification). - Issuer-based instruments (sustainability-linked): - Direct channel: interest-rate link to sustainability targets provides incentive, but effectiveness depends on the ambition and clarity of targets and additionality concerns. - Indirect channels: incentivize integration of sustainability objectives across the issuer’s agenda; issuer-level commitments may foster organizational changes. ### IX. Literature highlights and open questions - Empirical literature focused mainly on green bonds; mixed findings on pricing and environmental impact. - Key empirical findings cited: - Flammer (2021): green bond issuers improve environmental performance post issuance; certified issuances receive stronger investor interest and greater CO2 reductions. - Ehlers et al. (2020): find no clear evidence of lower or decreasing carbon intensity for green bond issuers; recommend complementing green bonds with issuer-level emissions ratings. - Tuhkanen and Vulturius (2020): many large European green bond issuers do not link green bond frameworks to climate targets; highlight shortcomings in post-issuance reporting raising greenwashing risks. - Unresolved issues: measurement of firm-level emission impacts, additionality, role of external reviews/certification for sustainability-linked instruments, and investor motivations. *Source: IMF staff (content unit: Appendix 6: Sustainability-linked Loan Borrowers vs Non-borrowers by Industry, from wpiea2021194-print-pdf).* ### Section I presented the limited literature on the environmental impact of green bonds. ### wpiea2021194-print-pdf - Section I presented the limited literature on the environmental impact of green bonds. ### Literature on environmental impact of green bonds - Flammer (2021) finds that green bond issuers improve their environmental performance more than other firms following the issuance of a green bond. - Results in Ehlers et al. (2020) and Tuhkanen and Vulturius (2020) are less conclusive. - Existing research is focused on green bonds; analysis of bank-based green products (green loans and sustainability-linked loans) and issuer-based instruments (sustainability-linked bonds and loans) has been hampered by data constraints. ### Channels through which green debt could affect environmental performance - Direct channel (additionality): - Defined as funding environmentally valuable projects that would not be financed without green debt. - Given the lack of evidence for a financially meaningful premium on green debt (see section II b), it is unlikely that projects become financially viable because they are funded with green debt. - Maltais and Nykvist (2020) issuer survey supports limited additionality. - Flammer (2021) notes green bonds are often small relative to issuer size, making large project-driven improvements unlikely. - Sustainability-linked instruments can provide direct financial incentives via interest rate adjustment tied to environmental targets; example: a recent sustainability-linked bond issued by ENI includes a 25 bps interest rate step up if carbon footprint and renewable capacity targets are not reached. - Indirect channels (likely more important): - Engagement with green debt mainstreams green and climate considerations in financial and corporate sectors. - Issuance process requires meeting green debt requirements and building internal capacity, facilitating organizational learning. - Guidelines and standards advance definitions of green assets/projects and pathways for firms to become climate friendly. - Green debt can credibly signal green credentials, generating positive spillbacks: - Lowering overall cost of capital for green firms (consistent with positive stock market reactions to green bond issuances reported by Tang and Zhang (2020) and Flammer (2021)). - Positive effects on employee, customer, and stakeholder relationships that may encourage further green investments. - Caveat: observed associations could reflect selection (already green firms choosing to issue green debt) rather than causal benefits. ### III. Sustainable and green debt market developments (key statistics) - Sustainable debt issuance reached $730bn in 2020. - Total global debt at $281tn (Institute of International Finance) — sustainable debt remains a small fraction. - Activity-based instruments accounted for more than 80 percent of sustainable debt issued in 2020. - Issuer-based instruments (sustainability-linked loans and bonds) reached a share of about 20 percent of sustainable debt in 2020 from close to zero in 2016. - Regional shares in 2020: - Europe accounted for more than half of sustainable debt issuance. - Western Hemisphere at 20 percent. - Asia at 15 percent. - Green bonds: - Largest sustainable debt category with about $300bn raised in 2020. - European issuers accounted for 55% of issuance. - EUR and USD combined account for 80 percent of issuance. - Corporates account for slightly more than half of all green bonds issued; financials and utilities are main corporate issuers. - National governments account for about 22 percent; asset-backed securities (ABS) account for 10 percent; supranational issuers 9 percent; US municipal debt 5 percent. - Green loans: - Volume in 2020 about $80bn, down from $93bn in 2019. - Almost exclusively used by corporates, mainly energy and utilities companies. - Sustainability-linked loans and bonds: - About $120bn of sustainability-linked loans in 2020, from $140bn in 2019. - Sustainability-linked bonds: total issuance in 2020 was $10.6bn. - ICMA published formal guidelines for sustainability-linked loans in June 2020 and sustainability-linked bond principles in June 2020. - Tanzania became the first sovereign borrower with a sustainability linkage in 2020. - Enel issued the first corporate sustainability-linked bond in 2019. ### IV. Greenness of issuers — data, measures, and findings - Greenness measurement: - Measured by CO2 emissions scaled by revenue and assets. - Emission reporting scopes: scope 1 (direct), scope 2 (indirect from purchased energy), scope 3 (other indirect value-chain emissions). - Presented data for scope 1 and 2 emissions; results robust to inclusion of scope 3. - Dataset and coverage: - Includes all firms in the S&P Global 1200 index. - Among S&P Global 1200 firms, 76% report at least scope 1 and 2 emission data. - CO2 emissions from Bloomberg and Thomson Reuters; revenues/assets from Bloomberg; green debt data from BNEF; environmental scores from Refinitiv via Thomson Reuters. - Observations: 148 unique firm-level observations for green bond issuers, 49 observations for green loan borrowers, and 70 observations for sustainability-linked loan borrowers. - Summary statistics (selected): - Green bond issuers are larger than average S&P Global 1200 firms by revenues and especially by assets. - Over half of green bond issuers are financials or utilities. - Green loan borrowers tend to be larger than average firms but less asset-heavy than green bond issuers; dominated by utilities (45%), followed by financials (20%). - Sustainability-linked loan borrowers: similar in revenues to average index firm and smaller by assets; industry structure dominated by utilities (21%), industrials (20%), and materials (17%). - A. Green Bonds — findings: - Green bond issuers have lower CO2 emissions relative to revenue and assets than firms that do not issue green bonds. - Green bond issuers are more likely to report CO2 emissions. - Results robust to emission scope and persist within industries with at least 10 observations (financials, utilities, industrials, consumer discretionary). - Green bond issuers tend to have higher environmental rating scores than non-issuers. - Green bond issuers that engage external reviewers for assurance tend to have lower CO2 emission intensities than those without external assurance. - Evolution around first issuance: - Emission intensities reduce notably in the second year after first issuing a green bond. - Significant reduction in the year of issuance relative to the previous year. - Interpretation: focused emission reductions around issuance year and continued improvement post issuance. - B. Green Loans — findings: - Green loan borrowers tend to have higher CO2 emissions relative to revenue and assets than firms that do not borrow with green loans; driven by industry composition (utilities overrepresented). - Within industries, no clear difference in emission intensities between green loan borrowers and other firms. - Evolution around loan origination: - Improvement from two years before to two years after origination, primarily due to a drop in emission intensities in the second year following the loan. - Overall, smaller reduction over the 5-year window compared with green bond issuers. - Possible explanations: - Definition of green loan less clear than green bonds; BNEF classifies earmarked loans as green irrespective of borrower self-labeling. - Green loan standards are relatively new and high level; external assurances are rare for green loans. - Green conditions in loans may be less ambitious and monitoring less stringent. - C. Sustainability-linked Loans — findings: - Sustainability-linked loans link interest rate to issuer-wide sustainability targets while loan proceeds can be used for general purposes. - Borrowers tend to have somewhat higher CO2 emission intensities in aggregate, driven by industry composition. - For utilities (21% of sample), sustainability-linked borrowers have somewhat lower emission intensities than other firms. - Evolution around loan: - Big improvement in CO2/revenue and CO2/assets in the year of the loan and some improvement in the following year. - Improvement larger than for green loans and similar magnitude as for green bonds. - Tentative interpretation: sustainability-linked loans may incentivize emissions reductions somewhat better than green loans; alternatively, greener firms may be more interested in sustainability-linked loans. - Data limitations: small sample sizes due to novelty; only three years of evolution possible given first loan in 2017; insufficient observations to analyze sustainability-linked bonds (exist since 2019). ### V. Conclusion — synthesis of findings and research needs - Market dynamics: - Green debt markets are rapidly evolving with growth in issuance, security design development, and refinement of governance, standards, and guidelines. - Green bonds remain the largest sustainable debt category; issuer-based instruments linking interest rates to firm-level sustainability targets have emerged. - Role in policy mix: - Many policymakers and investors view green debt as an important component to achieve the transition to a low carbon economy and ensure pricing of climate risks. - Empirical contributions and interpretation: - Documented CO2 emission intensity of green debt issuers relative to other firms and over time. - Findings: - Green bond issuers have lower emission intensities than other firms. - Green loan and sustainability-linked loan borrowers are in line with other firms after controlling for industry composition. - Green bond, green loan, and sustainability-linked loan borrowers lower their emission intensity over time at a faster rate than other firms. - Likely interpretation: issuers pursue green debt to signal green credentials (as argued by Flammer (2021)); engagement with investors, organizational learning, and mainstreaming green considerations also potentially contribute. - Direct financial benefits through lower debt cost are unlikely to be a major factor given lack of conclusive evidence of substantive pricing differences between green and conventional debt. - Research needs: - More research on environmental impact and its channels, motivations of market participants, and design of green debt to incentivize environmental objectives. - Comparative analysis of activity-based vs issuer-based instruments and of market-based versus bank-based debt financing as more data become available. *Source: wpiea2021194-print-pdf - Section I presented the limited literature on the environmental impact of green bonds.* ### References ### References ### Academic and policy references - Baker, M., Bergstresser, D., Serafeim, G. & Wurgler, J., (2018). Financing the Response to Climate Change: The Pricing and Ownership of US Green Bonds. National Bureau of Economic Research WP No. 25194. - BloombergNEF, (2020). 2H 2020 Sustainable Finance Market Outlook. - Cheong, C., & Choi, J. (2020). Green bonds: a survey. Journal of Derivatives and Quantitative Studies: Seonmul yeon’gu. - De Haas, R., & Popov, A (2021). Finance and Green Growth. CEPR Discussion Paper DP14012. - Deschryver, P., & De Mariz, F. (2020). What future for the green bond market? How can policymakers, companies, and investors unlock the potential of the green bond market?. Journal of risk and Financial Management, 13(3), 61. - Doronzo, R., Siracusa, V., & Antonelli, S. (2021). Green Bonds: the Sovereign Issuers' Perspective (No. 3). Bank of Italy, Directorate General for Markets and Payment System. - Ehlers, T., Mojon, B., & Packer, F., (2020). Green Bonds and Carbon Emissions: Exploring the Case for a Rating System at the Firm Level. BIS Quarterly Review September 2020. - Ehlers, T. & Packer, F., (2017). Green Bond Finance and Certification. BIS Quarterly Review September 2017. - European Commission, (2018). Action Plan: Financing Sustainable Growth. Communication from the Commission to the European Parliament, the European Council, the Council, the European Central Bank, The European Economic and Social Committee and the Committee of the Regions. - EU Technical Expert Group on Sustainable Finance, (2019). Report on EU Green Bond Standard. TEG Report, Proposal for an EU Green Bond Standard. - Flammer, C., (2021). Corporate green bonds. Forthcoming in Journal of Financial Economics. - Grippa, P., Schmittmann, J., & Suntheim, F. (2019). Climate Change and Financial Risk. Finance & Development, December 2019, Vol . 56, No. 4 . - Kapraun, J., & Scheins, C. (2019, March). (In)-Credibly Green: Which Bonds Trade at a Green Bond Premium?. In Proceedings of Paris December 2019 Finance Meeting EUROFIDAI-ESSEC. - Karpf, A., & Mandel, A. (2017). Does it pay to be green?. Available at SSRN 2923484. - Larcker, D. F., & Watts, E. M. (2020). Where's the greenium?. Journal of Accounting and Economics, 69(2-3), 101312. - IMF, (2019). Sustainable finance: Looking farther. Global Financial Stability Report October 2019. - Maltais, A., & Nykvist, B. (2020). Understanding the role of green bonds in advancing sustainability. Journal of Sustainable Finance & Investment, 1-20. - Park, S. K. (2018). Investors as regulators: Green bonds and the governance challenges of the sustainable finance revolution. Stan. J. Int'l L., 54, 1. - Schumacher, K. (2020). The Shape of Green Fixed Income Investing to Come. The Journal of Environmental Investing, 10(1). - Shishlov, I., Morel, R., & Cochran, I. (2016). Beyond transparency: unlocking the full potential of green bonds. Institute for Climate Economics, 1-28. - Tang, D. Y., & Zhang, Y. (2020). Do shareholders benefit from green bonds?. Journal of Corporate Finance, 61, 101427. - Tolliver, C., Keeley, A. R., & Managi, S. (2019). Green bonds for the Paris agreement and sustainable development goals. Environmental Research Letters, 14(6), 064009. - Tuhkanen, H., & Vulturius, G. (2020). Are green bonds funding the transition? Investigating the link between companies’ climate targets and green debt financing. Journal of Sustainable Finance & Investment, 1-23. - Winegarden, W. (2019). Environmental, Social, and Governance (ESG) Investing: An Evaluation of the Evidence. Pacific Research Institute. Available at: https://www. pacificresearch. org/wp-content/uploads/2019/05/ESG_Funds_F_web.pdf. - Zerbib, O. D. (2019). The effect of pro-environmental preferences on bond prices: Evidence from green bonds. Journal of Banking & Finance, 98, 39-60. ### Appendices and figures listed in the References section - APPENDIX 1: GREEN BOND ISSUERS VS NON-ISSUERS BY SCOPE OF EMISSIONS - Figure 25. Green Bond Issuers vs Non-issuers: Share of Firms by CO2 (Scope 1)/Revenue Buckets %, Financial Year 2019 - Figure 26. Green Bond Issuers vs Non-issuers: Share of Firms by CO2 (Scope 1)/Assets Buckets %, Financial Year 2019 - Figure 27. Green Bond Issuers vs Non-issuers: Share of Firms by CO2 (Scope 1+2+3)/Revenue Buckets %, Financial Year 2019 - Figure 28. Green Bond Issuers vs Non-issuers: Share of Firms by CO2 (Scope 1+2+3)/Assets Buckets %, Financial Year 2019 - APPENDIX 2: GREEN BOND ISSUERS VS NON-ISSUERS BY INDUSTRY - Figure 29. Financials: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - Figure 30. Utilities: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - Figure 31. Industrials: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - Figure 32. Consumer Discretionary: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - APPENDIX 3: GREEN LOAN BORROWERS VS NON-BORROWERS BY SCOPE OF EMISSIONS - Figure 33. Green Loan Borrowers vs Non-borrowers: Share of Firms by CO2 (Scope 1)/Revenue Buckets %, Financial Year 2019 - Figure 34. Green Loan Borrowers vs Non-borrowers: Share of Firms by CO2 (Scope 1)/Assets Buckets %, Financial Year 2019 - Figure 35. Green Loan Borrowers vs Non-borrowers: Share of Firms by CO2 (Scope 1+2+3)/Revenue Buckets %, Financial Year 2019 - Figure 36. Green Loan Borrowers vs Non-borrowers: Share of Firms by CO2 (Scope 1+2+3)/Assets Buckets %, Financial Year 2019 - APPENDIX 4: GREEN LOAN BORROWERS VS NON-BORROWERS BY INDUSTRY - Figure 37. Utilities: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - Figure 38. Financials: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - Figure 39. Consumer Discretionary: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - Figure 40. Industrials: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - APPENDIX 5: SUSTAINABILITY-LINKED LOAN BORROWERS VS NON-BORROWERS BY SCOPE OF EMISSIONS - Figure 41. Sustainability-linked Loans Borrowers vs Non-borrowers: Share of Firms by CO2 (Scope 1)/Revenue Buckets %, Financial Year 2019 - Figure 42. Sustainability-linked Loans Borrowers vs Non-borrowers: Share of Firms by CO2 (Scope 1)/Assets Buckets %, Financial Year 2019 - Figure 43. Sustainability-linked Loans Borrowers vs Non-borrowers: Share of Firms by CO2 (Scope 1+2+3)/Revenue Buckets %, Financial Year 2019 - Figure 44. Sustainability-linked Loans Borrowers vs Non-borrowers: Share of Firms by CO2 (Scope 1+2+3)/Assets Buckets %, Financial Year 2019 - APPENDIX 6: SUSTAINABILITY-LINKED LOAN BORROWERS VS NON-BORROWERS BY INDUSTRY - Figure 45. Utilities: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - Figure 46. Industrials: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - Figure 47. Consumer Discretionary: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 - Figure 48. Materials: Share of Firms by CO2 (Scope 1+2)/Revenue Buckets %, Financial Year 2019 *Source: wpiea2021194-print-pdf - References* --- _Source: https://www.imf.org/-/media/files/publications/wp/2021/english/wpiea2021194-print-pdf.pdf_