Global Debt Report 2024

3. Sustainable bonds: State of the market and policy considerations

Abstract

At the end of 2023, the outstanding amount of sustainable bonds issued by the corporate and official sectors totaled USD 2.3 trillion and USD 2.0 trillion, respectively. Green bonds dominate issuance in both sectors. The key difference between the sectors is that sustainability‑linked bonds (SLBs) played a lesser role in the official sector, with only USD 9 billion issued in 2023. Additionally, sustainable bonds constituted a modest portion of funding for central governments, representing merely 0.4% of all sovereign bond issuances in 2023.

There is no statistically significant evidence that companies systematically benefit from a premium for issuing a sustainable bond. This is a sign that market practices and regulations may need to change for sustainable bonds to fulfil their potential. For example, sustainable bond contracts typically allow issuers to refinance existing projects with the proceeds, rather than invest in new projects. At the same time, the share of sustainable bonds being assured by second party opinion providers has grown from less than half in 2019 to nearly three-quarters in 2023.

3.1. Introduction

This chapter focuses on the key characteristics and trends in the markets of sustainable bonds. This introduction presents a definition of sustainable bonds, and the box below summarises the key findings in the chapter. The second section presents the trends in the issuance of sustainable bonds, their outstanding value, their investor base, assets under management in sustainable investment funds, and the use of standards and taxonomies. The third section covers key issues for policy makers, including the main goals of the largest investors in sustainable bonds, whether there is a premium for sustainable bonds, the roles of relevant service providers, key characteristics of sustainable bond contracts, and the development and liquidity of sustainable bond markets. The final section raises policy questions based on the data and analysis presented in the previous sections.

Sustainable bonds can be classified into two major categories (ICMA, 2022[1]). “Use of proceeds bonds” are bonds whose proceeds should be used to either partially or fully finance or re‑finance new or existing eligible green, social or sustainable projects. In the case of “use of proceeds bonds” issued by financial institutions, the proceeds are typically allocated to finance or refinance the provision of loans for the development of eligible projects. “Sustainability-linked bonds” (SLBs) are bonds for which the issuer’s financing costs or other characteristics of the bond (e.g., its maturity) can vary depending on whether the issuer meets specific sustainability performance targets within a timeline, but whose proceeds do not need to be invested in projects with an expected positive environmental or social impact.

The “use of proceeds bonds” include green, social and sustainability bonds (GSS bonds). As their name suggests, the proceeds of green bonds must be applied to finance projects with expected environmental benefits, which may include, for instance, projects in renewable energy, clean transportation, biodiversity conservation, and wastewater management (ICMA, 2021[2]). In this classification, “blue bonds” and “climate bonds”, which focus on environmental issues related to the sea and climate change, respectively, would be classified as “green bonds”. The resources raised through social bonds must be invested in projects that aim to address or mitigate a specific social issue or seek to achieve positive social outcomes, including affordable housing, food security and the empowerment of minorities (ICMA, 2023[3]). Sustainability bonds are bonds where the proceeds should be used to finance a combination of both green and social eligible projects.

The following sections of this chapter analyse the key characteristics of both categories of sustainable bonds and their implications for companies, investors, and the entities comprising the official sector (the latter category includes national and subnational governments and their agencies, as well as multilateral institutions). However, one basic characteristic of sustainable bonds should be clear from the start. Sustainable bonds present the same rights and risks to their holders as with any other conventional bond, but they also create a commitment. In the case of the GSS bonds, the issuer ordinarily commits to having developed or to plan to invest in eligible sustainable projects with a value that is equal to or higher than the outstanding value of the GSS bonds it has issued. In the case of SLBs, the issuer commits to reaching sustainability performance targets, such as reducing its greenhouse gas (GHG) emissions. The potential for these commitments to effectively change the investment decisions of companies and the official sector, as well as whether investors assign any value to such commitments, are central to the chapter.

Key findings

• At the end of 2023, the outstanding amount of sustainable bonds issued by the corporate and the official sectors totalled, respectively, USD 2.3 trillion and USD 2.0 trillion. The total amount issued through sustainable bonds in the corporate and official sectors was six and seven times larger in 2019‑23 than in 2014‑18, respectively.

• Green bonds represented 74% of the total issued in corporate sustainable bonds in 2023, and SLBs were the second most relevant type of sustainable bonds with 10%. The issuance of sustainable bonds by non‑financial companies (USD 230 billion) represented 12% of all bonds issued by the non‑financial corporate sector, while the issuance by financial companies (USD 253 billion) represented 8% of the issuance of all bonds issued by the financial sector.

• The official sector has also been issuing a significant amount in sustainable bonds (USD 394 billion in 2023), but two significant differences are noticeable compared to the corporate sector. First, SLBs are much less relevant than GSS bonds for the official sector, with only USD 9 billion in issuance in 2023. Second, sustainable bonds remain a relatively limited source of funding for central governments, representing only 0.4% of the total issuance of all sovereign bonds in 2023.

• The standards developed by the International Capital Market Association (ICMA) were used to classify bonds as sustainable in issuances representing 93% of the total sustainable bond issuance in 2023. Issuers may typically refer to third‑party taxonomies for sustainable activities or to their own classification for eligible projects. The most‑often used taxonomies in 2023 were the ones from the Climate Bonds Initiative (66%), the European Union (14%), and the People’s Republic of China’s central bank and securities regulator (12%).

• There is no statistically significant evidence that companies systematically benefit from a premium for issuing a sustainable bond. Based on a limited sample of matched bonds, this also seems to hold true for the official sector.

• The share of sustainable bonds being assured by second party opinion providers has grown from less than half in 2019 to three-quarters in 2023. These service providers verify whether the bond contract is aligned with a specific sustainable bond standard.

• Two-thirds of the prospectuses mention that the refinancing of existing eligible projects with the proceeds is allowed, and no prospectus in the same sample explicitly mentions that the proceeds would not be used for refinancing in a sample of GSS bonds. Additionally, no GSS bond prospectus in the sample refers to a contractual penalty in case the issuer does not use all proceeds to finance or refinance eligible projects.

• There is limited evidence that sovereign issuance of green bonds has supported the development of the corporate green bond market. Still, sovereign issuers may have other reasons to issue sustainable bonds, including the diversification of their investor base.

3.2. Trends

3.2.1. Issuance and outstanding amount

Over the past five years, sustainable bonds have experienced a noteworthy growth as a source of capital market financing for both the corporate and official sectors. The total amount issued through corporate sustainable bonds was six times larger in 2019‑23 than in 2014‑18. Similarly, the amount issued by the official sector in the last five years was seven times larger compared to 2014‑18. Moreover, in 2021 a record amount of USD 713 billion was issued by corporates, of which 57% was issued by non‑financial companies (Figure ‎3.1, Panel A). Concerning the sustainable bonds issued by the official sector, they displayed record amounts in 2020 and 2021, with a total amount issued of USD 461 billion and USD 580 billion, respectively (Figure ‎3.1, Panel B). In 2020, multilateral institutions issued almost USD 230 billion in sustainable bonds. In 2023, corporate sustainable bond issuance decreased by 32% compared to 2021, while the official sector also saw the same decrease in its issuance of sustainable bonds (see sustainable bonds issuance over the total amount of bonds issued in Figure ‎3.8 and Figure ‎3.12).1

Figure ‎3.1. Global sustainable bond issuance by issuer type

Note: In Panel B, agencies and local governments include national government agencies (e.g., KFW), local governments (e.g., Prefecture of Shiga), and national development banks (e.g., Brazilian Development Bank). The category Multilateral Institutions includes organisations formed by three or more jurisdictions (e.g., International Finance Corporation), and the European Union.

Source: OECD Corporate Sustainability dataset, LSEG.

In 2023, the outstanding amount of sustainable bonds issued by the corporate sector totalled USD 2.3 trillion against USD 1.99 trillion by the official sector. The outstanding amount of sustainable bonds issued by the non‑financial corporate sector accounted for USD 1 230 billion, representing 8% of the outstanding amount of the bonds issued in this sector. Similarly, financial companies’ outstanding amount of corporate bonds totalled USD 1 068 billion, which is 6% of the outstanding amount of all corporate bonds issued by financial companies (Figure ‎3.2, Panel A). Moreover, the outstanding amount of sustainable bonds stood at USD 853 billion for “agencies and local governments” and at USD 686 billion for “multilateral institutions”, and for central governments this amount was comparably lower at USD 450 billion in 2023.

Figure ‎3.2. Outstanding sustainable bonds by issuer type

Source: OECD Corporate Sustainability dataset, LSEG.

Europe has been the most active region in the sustainable bonds market in both the corporate and official sectors. From 2014 to 2023, 45% of the global amount issued through corporate non‑financial sustainable bonds was raised by European companies. The People’s Republic of China (China) and the United States follow with 17% and 14% respectively (Figure ‎3.3, Panel A). Similarly, the issuance of sustainable bonds by financial corporates has been most relevant in Europe (54%), China (15%), Asia excl. China and Japan (10%) and the United States (7%). Central governments’ issuance of sustainable bonds has been done mainly by European jurisdictions, accounting for 67% of the global issuance of sustainable bonds from 2014 to 2023, followed by Latin American central governments with 16% of the global amount. Agencies and local governments’ sustainable bond issuance has also been dominated by European issuers (65% of the global amount), followed by agencies and local governments from Asia excl. China and Japan (13%) (Figure ‎3.3, Panel B).

Figure ‎3.3. Global sustainable bond issuance by region, 2014-23

Source: OECD Corporate Sustainability dataset, LSEG.

Prior to 2020, most of the issued corporate sustainable bonds were green, averaging 92% of all the amount issued via sustainable bonds each year (Figure ‎3.4, Panel A). In 2020, sustainability as well as social corporate bonds represented more than 10% each of the amount issued through sustainable corporate bonds. In 2022 and 2023, SLBs, sustainability and social bonds averaged 12%, 10% and 7% of the total amount of issued corporate sustainable bonds, respectively.

The issuance of green bonds is less prevalent in the official sector, representing 45% of the amount issued in the last three years on average. Social (29%) and sustainability (26%) bonds have been commonly used by governments and multilateral institutions over the last three years (Figure ‎3.4, Panel B). SLBs were issued for the first time by central governments and multilateral institutions in 2022, making up only 1% of the share of sustainable bonds issued in the last two years in the official sector.

Figure ‎3.4. Global sustainable bond issuance by type

Source: OECD Corporate Sustainability dataset, LSEG.

When analysing the issuer’s promised use of the proceeds as established in the GSS bond documentation, “renewable energy projects”, “energy efficiency” and “green construction/buildings” rank first among GSS bonds issued by financial and non‑financial corporates. “Eligible green projects”, which effectively present an open scope for investments, ranks fifth among GSS bonds issued by non‑financial corporates. In Table ‎3.1, the categories of “use of proceeds” in GSS bonds are ranked by their importance for each of the issuer types.

Agencies, local governments and multilateral institutions’ top priority when issuing GSS bonds has been “social expenditures”. Additionally, “biodiversity conservation”, “social expenditures” and “clean transport” rank among the top three priorities in the central governments’ issuance of GSS bonds, with 19%, 12% and 11% of the total amount raised, respectively. Notably, 10% of the amount raised via GSS bonds by multilateral institutions was issued with a general‑purpose use of the proceeds and, thus, not a sustainability‑related specific purpose.

Table ‎3.1. GSS bonds’ use of proceeds, 2014-23

	Corporate Sector		Official Sector
	Non-financials	Financials	Agencies and Local Governments	Central Governments	Multilateral Institutions
Agriculture	0%	0%	0%	1%	1%
Biodiversity Conservation	3%	4%	3%	19%	5%
Clean Transport	10%	11%	11%	11%	5%
Climate Change Adaptation	6%	6%	4%	10%	8%
Energy Efficiency	18%	15%	10%	11%	8%
Green Construction/Buildings	10%	17%	5%	7%	5%
Infrastructure	8%	4%	14%	3%	10%
Renewable Energy Projects	21%	15%	8%	7%	7%
Social Expenditures	3%	9%	28%	12%	25%
Sustainable Development Projects	1%	1%	2%	1%	7%
Waste Management	1%	2%	1%	4%	1%
Water or Wastewater management	4%	6%	4%	6%	4%
Eligible Green Projects	10%	7%	5%	6%	5%
General Purpose	5%	3%	4%	3%	10%

Source: OECD Corporate Sustainability dataset, LSEG.

Unlike GSS bonds, SLBs include specific sustainability performance targets to be met within a defined timeline. To this end, Key Performance Indicators (KPIs) are selected by the SLB issuer to achieve the sustainability objective at the entity level. As shown in Table ‎3.2, most of the KPIs in SLBs relate to climate transition, such as “scope 1 and scope 2 GHG emissions”, accounting for 28% of the SLBs issued by corporates. Other prominent KPIs related to the climate transition in corporate SLBs are, for example, “carbon and emissions intensity”, and “renewable energy”. Although less frequently, non‑climate transition key performance indicators were also mentioned for some corporate SLBs. These include, for instance, an “increase in women board members” (1.6%), an “increase in patient access to health treatments” (1.4%), and “racial and ethnic diversity” (0.9%). The KPIs of SLBs issued by the official sector are also included in Table ‎3.2, but only very few have so far been issued, which limits the comparability between the numbers of the corporate and official sectors.

Table ‎3.2. Key Performance Indicators (KPIs) in SLBs, 2019-23

	Corporate Sector	Official Sector
Carbon and Emissions Intensity	24%	24%
Energy Consumption and Efficiency	6%	10%
Increase in Women Board Members	2%	25%
Renewable Energy	9%	13%
Scope 1 and Scope 2 GHG Emissions	28%	8%
Scope 1, Scope 2 and Scope 3 GHG Emissions	3%	0%
Scope 3 GHG Emissions	7%	0%
Sustainable Forest Management	0%	6%
Other	21%	13%

Source: OECD Corporate Sustainability dataset, LSEG.

Almost 80% of sustainable corporate bonds from 2014 to 2023 have been issued with a medium‑term maturity – ranging from 2 to 10 years. Short‑term corporate sustainable bonds (with a maturity of less than two years) have represented only a fractional share over the years, accounting for only 3% of total corporate sustainable bonds. The official sector has issued sustainable bonds with longer maturities than the ones issued by the corporate sector. Indeed, while in 2023 long‑term bonds accounted for 19% of the total amount issued by the corporate sector, long‑term bonds by the official sector amounted to 37% (Figure ‎3.5).

Figure ‎3.5. Short, medium, and long-term sustainable bonds

Source: OECD Corporate Sustainability dataset, LSEG.

Sustainable bonds issued by corporates and by governments and multilateral institutions display less stable value‑weighted maturities when compared to all conventional and sustainable bonds during the past decade. On average, sustainable bonds issued by non‑financial corporates present a value‑weighted maturity of 11.9 years against a maturity of 9.4 years for all bonds issued by non‑financial corporates (Figure ‎3.6, Panel A). In contrast, sustainable bonds issued by financial corporates display on average a slightly shorter value‑weighted maturity of 6.1 years against 6.9 years for all bonds issued by financial corporates.

Multilateral institutions show similar maturities both for sustainable (7.9 years) and all bonds (8 years), although agencies and local governments’ value‑weighted maturity averages 8.3 years for sustainable and 9.6 years for all bonds (Figure ‎3.6, Panel B). In the case of the central governments, since 2017 the value‑weighted maturity of the issued sustainable bonds has been almost twice as long as the one of all issued bonds, averaging 16.8 years against 9.1 years, respectively, within the same period.

Figure ‎3.6. Value-weighted average maturity

Note: Bonds with maturities less than 1‑year maturity are excluded.

Source: OECD Corporate Sustainability dataset, OECD Capital Market Series dataset, LSEG.

A large share of SLBs include a call option that gives the issuer an option to redeem the bond before the end of its maturity. Over the 2014-23 period, while 45% of the amount issued via GSS bonds by the corporate sector included a call option, 80% of the SLBs issued by corporates included such an option (Figure ‎3.7, Panel A). A similar trend is also visible for the official sector. While 59% of the amount issued by official sector entities in SLBs included a call option, only 4% included such an option via GSS bonds. The existence of a call option in SLBs may raise a concern that issuers could seek to reduce the amount of any penalty that could arise from not meeting the sustainability performance targets by exercising the call option. The establishment of a penalty if targets are not met at the time of the exercise of the call option can potentially moderate that concern. Ul Haq and Doumbia (2022[4]) analysed 40 SLBs with a call option (up to December 2021) and found that 42.5% mentioned a penalty if the targets are not met at the time of the call.

Figure ‎3.7. Callable GSS bonds vs SLBs, 2014-23

Source: OECD Corporate Sustainability dataset, LSEG.

3.2.2. Issuer characteristics

Corporate sector

In 2014, sustainable bonds accounted for only 0.6% of the total amount issued by all non‑financial companies, yet in 2023 this ratio reached 11.7%, after having peaked at 16% in 2022 (Figure ‎3.8, Panel A). This rising trend has also been visible for corporate bonds issued by financial companies, whose equivalent ratio surged from 0.2% in 2014 to 8.1% in 2023. In 2023, every category of sustainable bonds experienced a drop in issuance with the exception of social bonds issued by financial companies, which presented a 23% increase from the year before (Figure ‎3.8, Panel B). Moreover, despite a decrease of 27% in SLBs issued by non‑financial companies in 2023 in comparison to the previous year, their share still accounted for 20% of the amount issued by non‑financials in 2023.

Figure ‎3.8. Sustainable bond issuance by corporates

Note: The black dots correspond to the share of sustainable corporate bonds over all corporate bonds.

Source: OECD Corporate Sustainability dataset, OECD Capital Market Series dataset, LSEG.

Globally, sustainable non‑financial (financial) corporate bonds accounted for 10% (7%) of all corporate bond issuance over the 2019‑23 period. Sustainable bonds depict a larger relative importance when compared to all corporate bond issuance in some regions, including Japan, Latin America, Europe, and Others. Notably, in China, the United States and Other advanced economies, sustainable bonds have represented 7% or less of all non‑financials and 4% or less of all financials (Figure ‎3.9).

Figure ‎3.9. Relative importance of sustainable bonds against all corporate bonds, by region, 2019‑23

Source: OECD Corporate Sustainability dataset, OECD Capital Market Series dataset, LSEG.

In 2022‑23, unlisted companies (i.e., companies that do not list their equity) issued about half of the sustainable bonds in both the non‑financial and financial corporate sector, following the same pattern in the issuance of conventional bonds. In the non‑financial corporate sector, unlisted companies issued 53% of all sustainable corporate bonds, followed by large listed companies that issued 26% of the amount, and smaller listed companies with 14% (Figure ‎3.10, Panel A). Similarly, unlisted financial companies issued 48% of the sustainable corporate bonds, and 27% was issued by large companies (Figure ‎3.10, Panel B). Interestingly, sustainable bonds have been issued to a larger extent by unlisted subsidiaries with a listed parent than other unlisted companies when compared to the issuance of all corporate bonds. For instance, while 14% of the amount of non‑financial corporate bonds was issued by an unlisted subsidiary with a listed parent, 20% of the amount issued through non‑financial sustainable corporate bonds was issued by the same type of issuer.

Figure ‎3.10. Corporate issuance by listed and unlisted issuers during 2022-23

Note: The inclusion of a company in the MSCI World Index or in the MSCI Emerging Markets Index as of September 2023 is considered as a proxy for a listed company being large. Unlisted companies (i.e. companies that do not list their equity) were classified as either a subsidiary with a listed parent company or “other unlisted companies”.

Source: OECD Corporate Sustainability dataset, OECD Capital Market Series dataset, LSEG, MSCI.

Although financial companies represent nearly half of the global issuance amount in corporate sustainable bonds over 2019‑23, there are substantial regional differences (Figure ‎3.11, Panel A). Financial companies have been more important issuers in Asia excl. China and Japan (60%), as well as in Other advanced economies (63%), than in Latin America (14%), Japan (28%) or in the United States (34%). Panel B of Figure ‎3.11 shows the industry distribution of the issuers of sustainable bonds while excluding financial companies, which provide loans to companies in different industries.

Figure ‎3.11. . Industry distribution of sustainable bonds, 2019-23

Source: OECD Corporate Sustainability dataset, LSEG.

In 2019‑23, companies from the utilities industry have been the most important issuers of corporate sustainable bonds in nearly all regions except in China, Japan and Others (Figure ‎3.11), where industrials – in China and Japan – are the most active sector in the issuance of sustainable bonds and real estate – in Others – accounting for 40%, 52% and 36% of the amount issued (excluding financials), respectively. The same holds when comparing their relative importance against all corporate bonds. Indeed, sustainable bonds represented 23% of the total amount issued via corporate bonds issued by companies from the utilities industry in 2023, which is more than two times the ratio of 9% for all non‑financial corporates in the same year. Renewable energy companies’ issuance may explain, at least partially, the importance of the utilities industry in the sustainable bonds market (Table ‎3.3Table ‎3.3).

Table ‎3.3. Relative importance of sustainable bonds against all corporate bonds, by industry

	2019	2020	2021	2022	2023
Basic materials	4%	5%	18%	15%	14%
Consumer cyclicals	1%	2%	9%	11%	14%
Consumer non-cyclicals	2%	1%	11%	13%	6%
Energy	1%	1%	8%	12%	11%
Financials	5%	5%	9%	10%	8%
Healthcare	1%	2%	11%	6%	3%
Industrials	3%	4%	8%	10%	8%
Technology	2%	3%	8%	10%	9%
Utilities	17%	15%	37%	33%	23%

Note: “Real estate” has been merged with “Financials” for all years.

Source: OECD Corporate Sustainability dataset, OECD Capital Market Series dataset, LSEG.

Official sector

Agencies and local governments, central governments and multilateral institutions have increasingly made use of sustainable bonds, particularly since 2020. Nevertheless, in relative terms, the issuance of sustainable bonds is still of minor importance for central governments but has become quite significant for multilateral institutions (Figure ‎3.12). Central governments have been the least frequent issuers of sustainable bonds when compared to the issuance of all bonds. In 2023, only 0.4% of the amount issued corresponded to sustainable bonds. Agencies and local governments have been more regular users of sustainable bonds, as they averaged 10% of all issued bonds since 2020.

Recently, multilateral institutions have stood out as the issuer group that relies most heavily on sustainable bonds for their capital market funding when compared to both other official sector entities and the private sector. While up to 2017 sustainable bonds accounted for a maximum of 7% of the total amount issued through bonds by multilateral institutions, in 2020 they represented 54%. Since 2020, at least 32% of the bond issuance has corresponded to sustainable bonds (Figure ‎3.12, Panel C). Notably, among all multilateral institutions the International Bank for Reconstruction and Development has issued 37% of the sustainable bonds, followed by the European Union (22%) and the European Investment Bank (12%). The International Development Association, the Asian Development Bank and the Inter‑American Development Bank each accounted for 4% of the amount issued by multilateral institutions.

Figure ‎3.12. Sustainable bond issuance by the official sector

Note: The black dots correspond to the share of sustainable bonds over all bonds issued by the official sector.

Source: OECD Corporate Sustainability dataset, LSEG.

3.2.3. Investors

This section analyses a dataset that gathers information from publicly available sources on bond investments to compare the investor base of sustainable bonds with those of conventional bonds. Although this dataset covers all bonds, it does not cover all investors. In other words, it is biased towards large and especially institutional investors, who typically have to disclose their portfolios to meet regulatory requirements. Overall, it covers around one-fifth of the investor base for these assets. Due to this constraint, this section focuses mostly on examining patterns between sustainable and conventional bonds instead of making claims on the whole investor base for sustainable bonds. Box ‎3.1, at the end of this section, explores the primary market data for selected sovereigns, which covers the totality of their respective issuances for the selected bonds.

Figure ‎3.13 Panel A shows the composition of the publicly available investor base for all types of bonds and distinct types of sustainable bonds. It first reveals that asset managers, which manage assets on behalf of asset owners by mainly investing in marketable securities, are the main holders of sustainable bonds. This reflects not only the importance of these types of investors but also the fact that a larger share of asset managers discloses their portfolios and, thus, are over-represented in this sample. Secondly, the figure shows that there are a lot of similarities in the publicly available investor base across sustainable bond instrument types. For corporate GSS bonds and SLBs, the share of the publicly available investor base is composed mostly of asset managers (from 70% for green bonds to 73% for sustainability bonds) and banks (from 11% for sustainability bonds to 19% for green bonds). Likewise, for the official sector’s sustainable bonds, asset managers represent the largest share of the publicly available investor base (from 59% for green bonds to 73% for social) followed by banks (from 17% for SLBs to 25% for green bonds).

The main difference between the publicly available holders of sustainable and all bonds is in the share held by banks and governments, which include, in the latter case, central banks. Notably, banks account for 10 and 20 percentage points more of the share of the publicly available investor base for sustainable bonds compared to all bonds issued by corporations and governments, respectively (Figure ‎3.13, Panels B and C). Another large difference presented in Figure ‎3.13 (Panels B and C) is the share held by governments of bonds issued by the official sector, which is roughly 49 percentage points smaller for sustainable bonds. It is worth noting that the disparity in the investor base can be attributable to variations in the underlying issuers of different bond types. For example, within the realm of sovereigns, the three largest issuers of conventional bonds – the United States, Japan, and China – have not to date issued sustainable bonds, and these differences in the publicly available investor base could, in theory, be attributed to the difference in the issuers. Nevertheless, upon examining a matched sample of the top 20 issuers of sustainable bonds, comprising nine sovereigns and eleven corporations, the noted disparity remains evident, reinforcing that these differences may not be driven by the issuer but rather by investors’ preference.2

Figure ‎3.13. Publicly available investor base for sustainable bonds

Note: Values as of January 2024.

Source: Bloomberg and OECD calculations.

Box ‎3.1. The investor allocation of sovereign GSS and conventional bonds at issuance for selected countries

Given the fact that sovereigns disclose information about the investors of their syndications, analysing sovereign primary markets makes it possible to obtain information on allocation of issuance amount by investor type. This box contrasts primary markets of all the GSS bonds issued by syndication in France, Germany and Italy, with those of conventional ones selected to match them by maturity, type, and issuance date. GSS bonds analysed were issued from 2017 to 2023 (see ‎Annex 3.A for details).

Figure ‎3.14 compares the investor base by region and type of investor, revealing that, on average, GSS and conventional bonds have a similar regional investor base (Panel A). For both bonds, around 20% of their primary market buyers are domestic, roughly 70% are non-domestic from Europe, and 10% reside in other regions. French domestic investors have acquired relatively more GSS bonds than investors in French bonds from other countries. This is the opposite of Italy: non‑domestic investors from Europe buy relatively more Italian GSS bonds than Italian conventional ones.

The share of GSS bonds from France, Germany and Italy purchased by pension funds and insurance companies is approximately double their share of matched conventional bonds, whereas the share acquired by hedge funds and public entities is roughly half that of their conventional bond counterparts (Figure ‎3.14, Panel B). This finding for pension funds and insurance companies diverges from the one in Figure ‎3.13, which did not show significant differences for pension funds and insurance companies.

Figure ‎3.14. Investor base of GSS and conventional bonds by region and investor type

Note: Each GSS is paired with a conventional bond from the same country with the most similar maturity among those allocated by syndication since the issuance of the first national GSS bond. Italian bonds are Buono del Tesoro Poliennale (BTPs), French bonds are Obligation Assimilable au Trésor (OATs). Weighted averages by country and issue size are used. Other PE refers to other public entities.

Source: (Agence France Trésor, 2023[5]), (Bundesrepublik Deutschland, 2023[6]), (Italian Treasury Department, 2023[7]), OECD calculations.

3.2.4. Sustainable investment funds

Following Section ‎3.2.3 about the investors in sustainable bonds, this section provides insights into the role of sustainable bonds in the composition and strategy of ESG-labelled and climate‑labelled funds, with a particular focus on open-ended funds and ETFs.

Even though the understanding of what constitutes sustainable investment can differ between countries and change over time, the total volume of assets overseen by professional investors who incorporate ESG and climate risk factors into their portfolio selection and management has seen considerable growth (OECD, 2022[8]). Nevertheless, as shown in Panel A of Figure ‎3.15 below, the assets under management of open-ended funds and ETFs (collectively called “investment funds” in this chapter) labelled as “sustainable” still represent only a small fraction of total assets under management in the investment funds industry.

Figure ‎3.15. Assets under management in traditional funds vs. sustainable‑labelled funds

Note: In Panel A, five of the nine Morningstar categories based on asset classes under management are reported: Equity, Fixed Income, Multi‑allocation, Alternative and Commodities. See ‎Annex 3.A for detailed description of these categories. Values as of January 2024.

Source: OECD Corporate Sustainability dataset, Morningstar Direct, LSEG.

A relatively small portion of the sustainable investing landscape comprises investment funds self‑identified as sustainable funds through the inclusion of terms like "ESG", “climate” or "sustainable" in their names (see ‎Annex 3.A for a detailed methodology). These sustainable funds, which include five different fund categories – equity, fixed income, multi-allocation, alternative and commodities – accounted for USD 1.4 trillion in January 2024, or 2.8% of the total assets under management of investment funds globally (Figure ‎3.15, Panel A).

Looking specifically at the top 50 holdings of fixed income sustainable funds – that is sustainable funds with most of their portfolios invested in fixed income assets – it is possible to observe that only 31% of the value of all bonds is related to sustainable bonds (Figure ‎3.15, Panel C; see ‎Annex 3.A for the calculation). Most of these sustainable bonds are GSS bonds (particularly green bonds, accounting for 84% of the total), with few investments in SLBs to date (Figure ‎3.15, Panel B).

There are at least two ways to explain the relatively low share of sustainable bonds in the portfolios of sustainable fixed income funds. First, it could be the case that asset managers, adequately fulfilling their mandate to consider the risks of their portfolio, financial returns and the impact of their investments on society and the environment, preferred to buy conventional bonds issued by entities with good sustainability‑related practices instead of sustainable bonds. This would mean, in some cases, for instance, that asset managers diligently reviewed the prospectus of a sustainable bond but were not convinced about the potential sustainability-related impact of the issuance in the decision-making process of the issuer. The second explanation for the low share of sustainable bonds in the portfolios of sustainable fixed income funds would be a more negative one. It may be the case that asset managers can label their fixed income funds as “sustainable” without necessarily needing to consider the environmental and social impact of their investments. If the second explanation holds true, the low share of sustainable bonds in their assets under management would be evidence of the so-called “greenwashing” in the investment funds industry.

Arguably the most important incentive for asset managers to “green wash” their funds would be if they can charge higher fees simply for including the label “sustainable” or “ESG” in the fund names. According to OECD analysis on funds’ fixed management fees for fixed income investment funds, it is possible to observe that sustainable funds charge average annual fees higher than traditional funds for both passive and active funds (Figure ‎3.16, Panel A).

Figure ‎3.16. Management fixed fees: sustainable vs. traditional fixed income funds

Note: Panel B only displays funds from managers of at least one active and one passive fund in both sustainable and traditional funds categories. Values as of January 2024.

Source: OECD Corporate Sustainability dataset, Morningstar Direct.

To have a more granular view, Panel B in Figure ‎3.16 analyses the fixed fees charged by the 17 largest asset managers who manage at least one active and one passive fund for both sustainable and traditional fund types. Focusing only on these asset managers may reduce any bias that may be present in Panel A of the same figure, such as the existence of more assets under management in regions where management fees may be higher. The analysis in Panel B, however, confirms that asset managers charge on average higher fixed fees for sustainable funds, both for active and passive funds. For active funds the average management fixed fees for sustainable funds are 1.00% while they are 0.38% for traditional ones. For passive funds the average management fixed fees for sustainable funds are 0.44%%, or more than the double of the management fees for traditional funds (0.21%).

3.2.5. Standards and taxonomies

As presented in Section ‎3.2, sustainable bonds are a recent innovation in debt markets. It is only natural, therefore, that very few jurisdictions have a regulatory or self-regulatory framework for sustainable bonds (e.g., as China and Japan do). There are two complementary and often-used international standards to classify a bond as sustainable, which are developed by two private sector‑led institutions: the International Capital Market Association (ICMA) and the Climate Bonds Initiative (CBI) (see Figure ‎3.17). ICMA’s Executive Committee is composed of 24 organisations comprising an equal distribution between investors, issuers, and underwriters (ICMA, 2020[9]). CBI’s Board is comprised of associations representing institutional investors and environmental non-government organisations (CBI, 2023[10]).

ICMA published the first edition of its “Green Bond Principles” in 2014, and, in more recent years, its “Social Bond Principles”, “Sustainability Bond Guidelines” and “Sustainability-Linked Bond Principles” (“ICMA Use‑of-proceeds Principles” to refer to the first three and “ICMA Principles” to all of them) (ICMA, 2023[11]). The ICMA Sustainability-Linked Bond Principles have five core components:

1. the issuer must select KPIs that are relevant to the issuer’s core sustainability policies, measurable or quantifiable, externally verifiable, and able to be benchmarked;

2. the targeted KPIs should be ambitious and determined on a predefined timeline set before or concurrently with the issuance of the bond;

3. the bond contract should establish that its financial and/or structural characteristics will vary depending on whether the targeted KPIs have been met (e.g., a variation of the coupon);

4. the issuer should report, at least annually, on the performance of the selected KPIs and on any information enabling investors to monitor the level of ambition of the targeted KPIs; and

5. the issuer should disclose an assurance attestation of the selected KPIs’ performance.

The ICMA Use-of-proceeds Principles have four core components:

1. the proceeds of the bond must be used for eligible projects, which should be appropriately described in the legal documentation of the bond;

2. the issuer must communicate how it assesses whether a project is eligible;

3. the net proceeds of the issuance should be tracked to attest they have been used to invest in the eligible projects; and

4. the issuer should report annually on the use of proceeds until full allocation, as well as provide a brief description of the projects developed and their expected impact.

Notably, ICMA Use‑of‑proceeds Principles allow issuers to use proceeds for refinancing eligible projects but, if this is the case, it is recommended that issuers provide an estimate of the share of financing versus re‑financing. The ICMA Principles recommend issuers to appoint an external review provider to assess the alignment of the sustainable bond with the relevant core components before issuing the sustainable bond (known as a “second party opinion”) and to verify the allocation of the funds after the issuance, but issuers may claim the alignment even without these assessments.

CBI published the first edition of its Climate Bonds Standard (CBS) in 2012 and, in newer versions of the standard, it covers both green “use-of-proceeds” bonds and SLBs with climate-related KPIs (CBI, 2023[12]). The CBS current edition builds on the ICMA Principles and, notably, the CBS recommends how the pre‑ and post‑issuance assurance of the sustainable bond would take place. The CBI also approves firms that can provide assurance that the sustainable bond is compliant with the CBS. Likewise, the CBI certifies bonds as compliant with the CBS after receiving an application that includes the assurance of such compliance by an approved assurance provider.

Importantly, ICMA Principles – and other standards that follow closely their framework – provide only a suggested and non-exhaustive list of broadly defined eligible green and social project categories. Issuers may typically refer to third-party taxonomies for sustainable activities or to their own classification of which projects would be eligible. The taxonomies classify economic activities in a similar way that nomenclatures for national economic statistics and international trade do (e.g., the International Standard Industrial Classification), but taxonomies also aim at defining whether these activities are sustainable (and, in some cases, setting different levels of sustainability). For activities that may not be considered inherently sustainable, the taxonomy would normally set a level of environmental or social performance above which the activity would be reckoned as sustainable (e.g., the construction of a building certified as green according to a specific benchmark).

Figure ‎3.17. Sustainable bonds issuance following different standards and taxonomies

Note: The values displayed correspond to the total amount issued following different standards and taxonomies. The values indicate the standards or taxonomies a sustainable bond is conforming to or aligned with (as reported by the issuers). A single sustainable bond can exhibit compliance/alignment with one or more standards or taxonomies.

Source: OECD Corporate Sustainability dataset, LSEG.

There are many national and international taxonomies for sustainable activities as a result of both government and market-led initiatives, including, for instance, the “Green Bond Endorsed Project Catalogue – 2021” (“China Green Taxonomy” in the figure above) set by China’s central bank and securities regulator (The People's Bank of China, 2021[13]) and the “Climate Bonds Taxonomy” developed by the CBI (CBI, 2021[14]). Notably, the taxonomies typically apply both to the corporate and official sectors, regardless of the differences in their activities. While the CBI Climate Bonds Taxonomy has been the prevalent taxonomy (see Figure ‎3.17 above), the most well‑known taxonomy for sustainable activities is probably the one set by the European Union (Regulation EU 2020/852). It focuses on six environmental objectives: climate change mitigation; climate change adaption; sustainable use and protection of water and marine resources; transition to a circular economy; pollution prevention and control; and protection and restoration of biodiversity and ecosystems. For an economic activity to be considered environmentally sustainable, it should substantially contribute to at least one of the mentioned environmental objectives and do no significant harm to the other objectives, as well as to comply with minimum safeguards related to social and governance-related issues (e.g., ensuring the alignment with the OECD Guidelines for Multinational Enterprises on Responsible Business Conduct). The European Union is also one of the founding members of the “International Platform on Sustainable Finance” (IPSF, 2023[15]), which, among other goals, aims to improve the interoperability and comparability of the EU taxonomy and the ones established by other jurisdictions.

The greater use of new taxonomies for sustainable activities set by regional and national institutions may raise three concerns. First, comparability between sustainable bonds may be reduced if they follow taxonomies that are meaningfully different. Second, taxonomies focused on the needs of the corporate sector may not be easily used by official sector entities. Third, organised industry interests may be successful in securing the inclusion of their business activities in a national or regional taxonomy in a way that may not fully take account of scientific evidence and broader policy objectives. Initiatives to improve the interoperability and comparability of regional or national taxonomies, such as the one led by the International Platform on Sustainable Finance, should therefore be welcomed.

3.3. Key issues

3.3.1. Incentives for investors

Investors may have three main reasons for acquiring a sustainable bond instead of a conventional bond with similar characteristics. First, individual investors and clients of institutional investors may be concerned about the social and environmental impact of their investments, or face pressure from other relevant stakeholders to demonstrate such concern. Second, investors with well-diversified portfolios may consider how the externalities of the companies they invest in might affect their long‑term financial return. Third, there may be a public policy that incentivises investments in sustainable bonds.

Sustainability-conscious investors and stakeholders

Individual investors can buy a bond through a brokerage firm. Arguably in most cases, they will have neither the sophistication nor the time to assess the bond’s legal documentation and the issuer’s business in detail. In the same way that one of these investors may consult the credit rating of the issuance to assess the credit risk profile of the bond, a sustainability-conscious investor might prefer to invest in a bond labelled as “sustainable” trusting that it will have a better social and environmental impact than a conventional bond.

Individual investors may often prefer to invest in capital markets through a professional investor, such as an asset manager or a pension fund. In this case, the individual investor will ordinarily choose among several investment vehicles that allow managers different levels of discretion on how to select assets to invest in. The mandates in these vehicles can restrict, for instance, the asset classes managers can acquire and, more recently, some of them promise to consider environmental and social matters in the investment‑making process. As seen in Section ‎3.2.4, investment funds self-labelled as “sustainable” accounted for USD 1.3 trillion in January 2024, or 2.8% of the total assets under management of investment funds globally.

The precise obligations of asset managers and other professional investors when selecting bonds to invest in will largely depend on their contracts with their investors. In some cases, managers may need to simply follow an index composed of sustainable bonds and, therefore, they would not be able to buy conventional bonds. In most other cases, however, a mandate to consider both financial returns and the sustainability‑related impact of investments may not mean that asset managers can only acquire sustainable bonds for their fixed-income portfolio. For instance, investing in the conventional bonds issued by a company with a positive social and environmental impact may be well-aligned with the sustainability‑related goals of the asset managers. Nevertheless, it is undeniable that the label “sustainable” may be attractive for some asset managers with sustainability-related goals merely from a compliance perspective.

In a less direct way, insurance companies, banks and central governments may decide to invest in sustainable bonds to improve their reputation as “sustainable” or as a response to external pressure. For instance, insurance companies may invest part of their reserves in sustainable bonds in order to be able to advertise that it is a “sustainability‑conscious” institution and, therefore, possibly attract more clients, even if the insurer does not offer any product labelled as “sustainable”. Also as an example, banks may suffer pressure from civil society organisations if their credit portfolio is concentrated in high-polluting companies, which can arguably harm their reputation with some environment-conscious clients and employees. Without discussing the substance of its content, the title of the following report serves as an example of mentioned pressure from civil society: “Lethal Investments: The Health Consequences of Cash Flows into Coal – how major global banks are financing some of the deadliest coal plants in the US” (Sierra Club, 2023[16]).

Governments and central banks may also face pressure from civil society organisations to consider sustainability‑related matters in their asset allocation decisions because citizens are de facto the final beneficiaries of government’s holdings. Central banks, who are typically the main holders of securities in the official sector, acquire (especially sovereign) bonds for two main reasons: (i) due to central banks’ asset purchase programmes (i.e., quantitative easing); (ii) to establish foreign-currency reserves. In that regard, there is evidence that governments’ holdings of conventional and sustainable bonds differ in a relevant way (see Figure ‎3.13). Additionally, a minority of central banks have sustainability objectives in their mandates (Dikau and Ulrich, 2021[17]), which includes the ECB (European Central Bank, 2022[18]). Considerations to establish a sustainability objective for a central bank may include not only the environmental and social impact of its portfolio, but also the central bank’s example for institutional investors of good sustainability-related policies and practices.3

Considering the publicly available information, most central banks in G7 countries have not typically bought sustainable bonds as part of their asset purchase programmes (OECD analysis using central banks’ information on their websites). As central banks’ bond holdings are substantial (around USD 14 trillion in the G7 central banks as of 2023), the absence of some major central banks in the sustainable bonds market greatly affects its investor base compared to conventional bonds. Among G7 countries, the only central bank that holds a noticeable amount of sustainable bonds in its portfolio is the Eurosystem, which between 2018 and 2022 purchased an increasing amount of corporate and government sustainable bonds (Elderson and Schnabel, 2023[19]). Nevertheless, most central banks have discontinued their asset purchase programmes and are unwinding their balance sheets since the beginning of the monetary policy tightening cycle that started in 2021‑22.

Portfolio management

Even for investors that are not concerned with the social and environmental impact of their investee companies, a decision to invest in sustainable bonds may still maximise their financial return for a given level of risk. This would be true in two circumstances. First, where the investor wants the company to adopt a new business strategy that takes into better account long-term environmental and social trends such as climate change, and, therefore, a strategy that would maximise the company’s value. Second, where the investor wishes the company to reduce its negative externalities (or increase its positive ones) despite a possible reduction of the company’s value, but with the view that the benefit for the investor’s other investee companies would more than compensate the loss in value for the first company. An example of the second circumstance would be the reduction in GHG emissions by an investee major energy company that would facilitate the transition to a low carbon economy, which would be financially positive for investee companies in the tourism sector with assets in tropical regions.

In either of the abovementioned circumstances, the most important point within the scope of this chapter is to highlight that, at least potentially, the commitments in a sustainable bond contract can alter the decision-making process of a company in a more efficient way than, for instance, buying equity shares in the same company. For example, a SLB with ambitious targets for a company to reduce GHG emissions and a meaningful potential coupon increase in case the target is not met could be more effective in changing corporate behaviour than a minority equity stake where the investor would not be in a position to alter the company’s strategy.

Public policies

Some jurisdictions offer favourable regulation and financial incentives to encourage investments in sustainable bonds, including tax credit bonds, direct subsidy bonds, and tax-exempt bonds (CBI, 2022[20]). Tax credit bonds offer investors the opportunity to receive tax credits in lieu of traditional interest payments. Direct subsidy bonds are another option that provides government cash rebates to offset net interest payments. Tax-exempt bonds, for instance in US municipal bonds and Brazilian wind projects, allow investors to avoid income tax on interest.

Some central banks are also supporting sustainable investments to meet climate targets, often through favourable policies for sustainable bonds (European Central Bank, 2022[18]). These measures may include lower capital requirements or preferential treatment in risk-weighting assessments, making such bonds potentially more attractive to financial institutions.

Banks hold bonds, among other reasons, to meet liquid asset requirements and manage their short‑term liquidity. For these two objectives, banks use bonds as collateral for repurchase agreement (repo) transactions with other private institutions, the central bank or the country’s debt management office. Therefore, the eligibility of an asset for meeting liquid asset requirements and to be used for repo transactions is a major incentive for banks to hold a security. An analysis of the repo eligibility of sustainable bonds issued by both governments and corporates reveals that sustainable bonds are mostly eligible in Europe, which helps explain why European banks held 84% of all banks’ holdings in sustainable bonds that were publicly available in July 2023 (LSEG, OECD calculations). More specifically, of the 41 sovereign sustainable bonds that are repo eligible with central banks, 38 are eligible in European countries (32 eligible with the ECB, 4 in the United Kingdom and 2 in Hungary);4 of the 823 corporate sustainable bonds that are repo eligible with central banks, 750 are eligible with the ECB;5 and of the 501 agency and supranational sustainable bonds that are repo eligible with central banks, 463 are eligible in European countries (457 with the ECB, 4 in Sweden and 2 in Switzerland).6

Stewardship codes

The recently revised G20/OECD Principles of Corporate Governance recognise stewardship codes as a mechanism that may complement regulatory requirements to encourage institutional investors’ engagement with their investee companies (Principle III.A). As a matter of fact, stewardship codes are increasingly integrating sustainability considerations into the engagement and voting policies of institutional investors.

Japan’s Stewardship Code provides that institutional investors are responsible for enhancing their investee companies’ corporate value and sustainable growth, taking into consideration ESG factors (Financial Services Agency, 2020[21]). Similarly, the Brazilian Stewardship Code states in its third principle that institutional investors should integrate ESG factors in their investment processes and scrutinise their impact on the sustainable development of the securities’ issuers (Associação de Investidores no Mercado de Capitais, 2016[22]).

Stewardship consists of pursuing a long-term increase in investors’ returns together with environmental and societal benefits, according to the UK Stewardship Code. In addition, the UK code also recognises the significant increase in investments other than listed equity. In this respect, they include reporting expectations for fixed income investments such as the review of prospectus and transaction documents (Financial Reporting Council, 2020[23]).

Stewardship codes do not typically, however, have specific recommendations related to investments in sustainable bonds. For instance, the major existing stewardship codes do not highlight the importance of investors’ analysis on whether SLBs’ performance targets are ambitious. SLBs are a promising tool for aligning investors’ sustainability-related preferences with investee companies’ impact on the environment and society. However, an SLB with an unambitious target functions de facto is a conventional bond because it does not change the decision-making process of the issuer’s leadership.

3.3.2. Incentives for issuers

The growth of the sustainable bonds market might reflect investors' increasing focus on sustainable and responsible business issues, as an incentive to invest in these assets. However, at the same time, it is not clear if these assets provide economic incentives to issuers by trading at a premium, which is defined, for sustainable bonds, as ‘greenium’. In the following part of this section, an overview of existing literature about this concept and the results of an OECD analysis are shown.

The term 'greenium' or 'green premium' refers to the yield difference between a conventional bond and a green bond with similar characteristics (Bolton et al., 2022[24]). Essentially, a 'greenium' infers that the yield an investor accepts to invest in a 'green' asset is less than the yield the same investor would be willing to accept to invest in an equivalent conventional asset, meaning that the issuer of the bond can obtain financing at lower cost when issuing a green security. This differentiation manifests itself in the primary market as a higher price for the green bond compared to a conventional bond at issuance. The existence of a 'greenium' in secondary markets would imply that a green bond is being traded at a superior price – or a lower yield – compared to a conventional bond with similar traits. This indicates that an environmentally conscious investor is willing to receive a reduced yield in return for the chance to contribute to a greener alternative.

Literature on greenium in bond markets

In past studies, several methodologies and key variables have been used for possible identification of a greenium, with different and not always consistent results. The conclusions indicate varied outcomes concerning the presence of a green premium in the primary market, while showing a more consistent agreement in studies focused on the secondary market (MacAskill et al., 2021[25]).

A first strand of literature found mixed evidence about the existence of the greenium. Hinsche (2021[26]) conducted its studies employing differing methods and samples to check the existence of the greenium, discovering a negative premium confined to Next Generation EU (NGEU) bonds. Zerbib (2019[27]) evaluated the influence of pro‑environmental preferences on bond prices using a matching method, before employing a two‑step regression process to compare yield spreads of green and conventional bonds within the secondary market. Across a sample ranging from July 2013 to December 2017, a negative marginal premium of 2 basis points was found, primarily for financial companies and low‑rated bonds. Bolton et al. (2022[24]) reveal several key findings. Initially, they observe that the greenium, if present, is not immediately apparent within the data. Even with a robust attempt to identify it within a sample of 63 bond pairs issued by the same entity in the same currency and with similar maturity (within a year), their search proved unfruitful. In their second point, upon comparing “twin bonds” (those of identical maturity) issued by EU countries, they discern a statistically significant, albeit minor, greenium – within the range of single‑digit basis points (Bolton et al., 2022[24]). Evidence from the US municipal green bond market suggests the green premium is only a more recent development. A review of a large dataset of US municipal bonds indicated that, overall, the yield curve of green bonds systematically falls below conventional bond yields (Karpf and Mandel, 2018[28]). Finally, Ando et al. (2022[29]) discovered an average greenium of 3.7 basis points for EUR‑denominated government bonds and nearly 30 basis points for USD‑denominated government bonds.

In contrast, Elhers and Packer (2017[30]) found no price premiums in their secondary market analysis, although they did identify a price premium for green bonds in the primary market. Similarly, Tang and Zhang (2018[31]) found no price premium when analysing the yield spread between the same issuing company within the same year. However, when assessing a broader sample, they found that green bonds were issued at a yield spread 6.94 basis points lower than corporate bonds issued by similar firms. A recent study by Doronzo et al. (2021[32]) compared bond prices in France, Belgium, Ireland, and the Netherlands. Results showed that yields from sovereign green bonds were nearly identical to those of conventional government bonds, consistent with previous findings for corporate and municipal green bonds. The findings in another paper (ESMA, 2023[33]) do not support the presence of a uniform or consistent pricing benefit across any category of sustainable bonds. Moreover, the study reveals that the pricing premiums experienced by some issuers can be attributed more to their issuer characteristics than their contractual commitments to sustainability.

These mixed findings suggest that if the greenium does exist as an incentive for issuers of sustainable bonds, it is generally minimal – with possible exceptions for some countries that are greatly vulnerable to the effects of climate change, transition and physical risks (Bolton et al., 2022[24]).

An analysis of a premium for sustainable bonds

The results from an empirical analysis in this report are consistent with the existing literature. The premium for a bond being labelled as sustainable is not statistically significant and may depend on several variables that are not entirely related to the nature of sustainable or conventional bonds per se. Using a data sample consisting of 69 718 corporate bonds and 63 022 official sector bonds, of which 5 563 and 270 are sustainable bonds respectively, and following the methodology in Bolton et al. (2022[24]), it has been possible to obtain two sets of quasi‑exact matched bonds for both categories: 1 736 matched corporate bonds and 72 matched official sector bonds (see ‎Annex 3.B for the descriptive statistics and the complete methodology). In these two sets, each sustainable bond matches with one conventional bond in all of the following seven variables: issuer, currency, issue date, maturity year, coupon, coupon class (fixed vs. floating coupon) and asset‑linked security category (e.g., inflation‑linked security).

Matching sustainable and conventional bonds issued by the same entity is essential because differences between issuers can affect the characteristics of sustainable bonds in ways that are not immediately observable. The main result of the empirical work, consisting of an application of a t‑test and a basic linear regression model, found no statistically significant evidence of a premium, here expressed as the impact of being labelled as sustainable on the yield to maturity. However, for the official sector bonds the results must be interpreted with caution given the limited number of observations.

When looking at the difference in the yield to maturity between sustainable bonds and their matched conventional bonds, the trends do not show any particular and constant differences (Figure ‎3.18), but they do show a common trend with the difference in terms of yield of around 1 and 3 basis points (0.01% for official sector bonds, -0.03% for corporate bonds). Any differences, if they do exist, are neither statistically significant or consistent. Furthermore, when looking at the average premium calculated by currency and issuer’s domicile, not only the constant absence of a premium for “sustainability” is confirmed, but a potential common trend is also missing (Figure ‎3.20).

Looking at the averages, for the official sector, for the 2016‑19 period, the sustainable premium of 145 basis points (-1.45%) suggests that sustainable bonds had a meaningfully lower yield than conventional bonds. Conversely, for the 2020‑23 period, sustainable bonds had yield that was on average 1 basis point (0.102%) higher than comparable conventional bonds. In the corporate sector, the premium is even less pronounced. During the 2014‑18 period, corporate sustainable bond had a yield 6 basis points (0.06%) higher on average than comparable conventional bonds. This difference reduced to 4 basis points (0.038%) in the 2019‑23 period, signalling a trend towards yield parity between conventional and sustainable bonds.

Figure ‎3.18. Average difference in yield‑to‑maturity of conventional vs. sustainable bonds for matched corporate and official sector bonds

Note: Only pairs of matched bonds are considered. Matching procedures is based on an exact matching methodology (see ‎0) based on issuer, coupon type and currency, and the nearest matching for issue date, issue year, maturity year, and coupon.

Source: OECD Corporate Sustainability dataset, LSEG.

Figure ‎3.19. Yield‑to‑maturity of conventional vs. sustainable bonds for matched corporate and official sector bonds, 2014-23

Note: Only pairs of matched bonds are considered. Matching procedures is based on an exact matching methodology (see ‎0) based on issuer, coupon type and currency, and the nearest matching for issue date, issue year, maturity year, and coupon.

Source: OECD Corporate Sustainability dataset, LSEG.

Figure ‎3.20. Average difference in yield‑to‑maturity of conventional vs sustainable bonds

Note: Only pairs of matched bonds are considered. Matching procedures is based on an exact matching methodology (see ‎0) based on issuer, coupon type and currency, and the nearest matching for issue date, issue year, maturity year, and coupon.

Source: OECD Corporate Sustainability dataset, LSEG.

Trends in oversubscription

Large issuances of green bonds in EUR and USD during the past three years consistently showed slightly higher average oversubscription ratios (orders received over the value of the issuance) than conventional ones. The oversubscription ratios ranged from 2.6 to 5.4 for green bonds and from 2.3 to 3.3 for conventional bonds (Figure ‎3.21).

Figure ‎3.21. Average oversubscription of green and equivalent conventional bonds

Source: Climate Bond Initiative, (CBI, 2023[34]).

Integration of sustainability factors in public debt management

According to a survey held in 2022 with OECD members and accession countries (OECD, 2023[35]), central governments consider sustainability factors in debt management operations, which includes the issuance of sustainable bonds, for numerous reasons (Figure ‎3.22). The most common motivations, which were mentioned by approximately two‑thirds of these countries, are to support the market for sustainable finance instruments, align debt management operations with the government’s ESG policy, and the diversification of the investor base of sovereign debt. Other motivations include improving the accountability and transparency of the government’s sustainability and climate-related initiatives, and improving sovereign creditworthiness.

By exclusion, it is interesting to note that countries often do not issue sustainable bonds to minimise financing costs, as they do not observe a reliable greenium in sustainable bonds’ issuances. In addition, among the main motivations frequently mentioned, only one (diversification of the investor base) is directly related to debt management, while all others are focused on supporting sustainability through capital market development, transparency or supporting government sustainability policy.

Figure ‎3.22. Motivation to consider sustainability factors in debt management operations in OECD and accession countries

Source: 2022 OECD and accession countries Survey on Primary Market Developments (OECD, 2023[35])

3.3.3. Service providers

As in any bond issuance, the issuer typically needs to hire an underwriter to be responsible for the offering of the sustainable bonds and an independent trustee may be appointed to represent the interest of bondholders. Among other activities, underwriters may contact potential investors, fulfil filing obligations, and supervise whether the issuance complies with regulatory requirements. Trustees are commonly appointed when bonds will be offered to many investors, in which circumstance individual investors may not have the incentives to supervise whether the issuer is complying with all its obligations.

In the cases of both underwriters and trustees, there is no evidence their fees would be higher for sustainable bonds than they are for conventional bonds. In the OECD analysis presented in Section ‎3.3.4 below, there were two cases in which non‑financial companies each issued both SLBs and conventional bonds. In the cases of both companies, the underwriter fees were slightly lower for the SLBs’ offerings in comparison to the offerings of conventional bonds. This is unexpected, however, as, all other factors being equal, underwriters and trustees of sustainable bonds have a broader scope for their analyses and, therefore, a potential liability in the issuance of sustainable bonds. For instance, underwriters may need to assess whether a sustainable bond could effectively be labelled as such, and a trustee may be contractually required to check whether the issuer has complied with its disclosure obligations on the use of proceeds for GSS bonds. Both service providers may decide to charge more due to the greater scope of their activities and potential liability but, at least in the prospectuses analysed for this chapter (see a description of the sample in Section ‎3.3.4), underwriters often include disclaimers stating that there is a risk that the sustainable bond may not be considered as a “sustainable finance security” according to the relevant regulation for the investor. Moreover, it seems to be a relatively common practice to state that the “second party opinion” is not incorporated by reference into the prospectus, which may effectively mean that the underwriter is not liable if the “second party opinion” is not accurate.

What is clearly an extra cost for the issuance of sustainable bonds in comparison to the one of conventional bonds is the market practice of providing a second party opinion on whether the bond contract is aligned with a specific sustainable bond standard (the most common are the ICMA Principles, as analysed in Section ‎3.2.5) and/or a taxonomy for sustainable activities. The ICMA Principles recommend but do not require a second party opinion for issuers to claim alignment with the standard.

Figure ‎3.23. Sustainable bonds issuance with (without) the use of a second party opinion provider

Source: OECD Corporate Sustainability dataset, LSEG.

Sustainable bonds have increasingly been assured by second party opinion providers. While in 2019 47% of the sustainable corporate bonds used a second party opinion provider, in 2023 81% did so. The official sector has also used second party opinion providers in their issuances. In 2023, 83% and 82% of the sustainable bonds issued by, respectively, agencies and local governments, and by central governments, used second party opinion providers. In contrast, multilateral institutions have used to a lesser extent second party opinion providers as part of their sustainable issuance, with only 26% of their sustainable bonds issued in 2023 using a second party opinion (Figure ‎3.23). Globally, the service providers with the greatest number of second party opinions developed include firms that belong to groups widely known for providing indices, credit ratings and proxy advice. Only to a smaller extent do auditing firms provide second party opinions.

Some of the same firms that provide second party opinions and other service providers offer assurance reviews of other aspects of the sustainable bonds before and after the issuance. Pre‑issuance assurance can involve, for instance, a review of the processes followed to consider projects as eligible or an estimation of the expected social or environmental impact of the projects to be financed by the issuance. Post‑issuance assurance can include, for example, an assessment of the effective impact of the projects financed by the issuance of GSS bonds or a review of the issuer’s performance related to the KPIs selected in a SLB. As mentioned in Section ‎3.2.5, a post‑issuance assurance of the performance related to KPIs is a condition for an issuer to legitimately claim alignment with the ICMA Sustainability‑Linked Bond Principles.

Service providers that offer second party opinions, as well as pre- and post-issuance assurances, have some standards to guide their reviews. For instance, these standards include the International Standard on Assurance Engagements (ISAE) 3000 for assurance engagements other than audits or reviews of historical information, which is set by the International Auditing and Assurance Standards Board (the same organisation is currently developing the International Standard on Sustainability Assurance 5000). More specifically for sustainable bonds, there are also CBI’s Climate Bonds Standard and ICMA Guidelines for Green, Social and Sustainability Bonds External Reviews.

The same conflicts of interest that exist for external auditors and credit rating agencies are also present for providers of second party opinion and other forms of assurance for sustainable bonds. The bond issuers may hire these service providers for several sustainable bond issuances, but they provide a service that is relevant to the public interest. While most external auditors and some credit rating agencies are regulated and supervised, providers of second party opinion and other forms of assurance for sustainable bonds do not – with some exceptions – face the same scrutiny.

In the approved framework for an EU Green Bond Standard, all green bonds will need to be reviewed by an external services provider both for its compliance with the standard and the alignment of its eligible projects with the EU taxonomy for sustainable activities (European Commission, 2023[36]). Notably, the firms providing such second party opinion will need to be registered with and supervised by the European Securities Markets Authority (ESMA). Likewise, the Climate Bonds Initiative offers to register providers of assurance for sustainable bonds after a review of whether they “follow best practice guidelines concerning ethics, independence, management of conflicts of interest, competence, documentation, and quality controls” (CBI, 2023, p. 11[12]).

3.3.4. Contracts

There are at least two explanations for the fact that issuers of sustainable bonds do not systematically benefit from a premium for the “sustainability” aspect of the bonds as analysed in Section ‎3.3.2. First, investors may not meaningfully prioritise the environmental and social impact of their investments, and some key issues raised in Section ‎3.3.1 may indicate that this hypothesis might be true for some institutional investors. Second, there is the possibility that, while some investors may significantly consider the environmental and social impact of their investments, market practitioners may still need to improve how sustainable bond contracts are designed to fulfil investors’ expectations.

An analysis of a sample of 120 sustainable bonds issued between 2017 and 2022 can provide better clarity regarding the market practices for designing contracts. The sample is composed of the 60 largest issuances and 60 randomly selected issuances that have their prospectuses easily accessible in financial databases in English, and these issuances belong to the following categories: GSS bonds issued by financial corporations (23); GSS bonds issued by non‑financial corporations (24); SLBs issued by financial corporations (18); SLBs issued by non‑financial corporations (28); GSS bonds issued by official sector entities (23); SLBs issued by official sector entities (4). The sample includes issuers from jurisdictions such as Australia, Bermuda, Brazil, Canada, Chile, China, France, Hong Kong (China), Japan, Luxembourg, Mexico, New Zealand, South Africa, the United States, and Uruguay. The sample also includes multilateral institutions such as the African Development Bank and the European Union.

With respect to the “use of proceeds” of GSS bonds in the sample, three issues are worth noting: the possibility of refinancing; the existence of contractual penalties; and the commitment to provide annual assurance.

As allowed by the ICMA Use-of-proceeds Principles, two thirds (65%) of the GSS bonds’ prospectuses mention that the refinancing of existing eligible projects with the proceeds is allowed, but, differently from what the mentioned principles recommend, the prospectuses do not provide an estimate of the share of financing versus re‑financing. Notably, no prospectus in the sample specifically mentions that the proceeds would not be used for refinancing. The possibility of refinancing an eligible asset may incentivise the issuer not to sell it, but, evidently, no new investment will be made because the asset already exists. Actually, in some circumstances, it may even be negative for society to incentivise a company with access to the sustainable bond market to keep an asset instead of selling it to another company that may be able to operate the asset in a more efficient way but which does not have easy access to public capital markets (for instance, if the former is a listed company and the latter is not).

No GSS bond prospectus in the sample refers to a contractual penalty in case the issuer does not use all proceeds to finance or refinance eligible projects. As a matter of fact, the prospectus of a major non‑financial corporate issuance defines that 70% of the proceeds would be invested in eligible green projects but the remaining would be used as working capital of the issuer. In 37% of the issuances, the prospectus notes that the non-compliance with the commitment to use proceeds for eligible projects would not be considered an event for the default of the GSS bond. Of course, this does not mean that issuers can simply disregard their obligation to use the proceeds of the issuance according to what is defined in the bond contract. Moreover, while symptomatic of their lack of importance for market participants, some prospectuses may have not included penalties that are indeed established in the bond contract. However, leaving a lawsuit for a damages award as the only recourse to investors may not offer them enough safety in some jurisdictions, especially because the effective damage may be difficult to assess in such a case.

Still in relation to the GSS bonds in the sample, only half of the prospectuses establish that the issuer will provide an annual assurance of the “use of proceeds”. In two cases, the assurance of the “use of proceeds” will be published quarterly. If the chosen assurance provider is highly qualified and effectively independent from the issuer, the information could be valuable for investors. A smaller number of prospectuses (35%) establish that the issuer will provide an annual assurance – and, in two cases, a quarterly assurance – of the impact of the projects financed by the issuance of the GSS bonds (21 out of 60 issuances).

With regards to the SLBs in the sample, two issues are worth noting: the consequence of not reaching their sustainability performance targets; and the commitment to provide a report on the performance against the KPIs relevant to the targets.

All but three issuers in the sample face the same consequence if they do not meet the sustainability performance target(s) set in their SLB contract: an increase in the annual coupon rate after the predetermined date to reach the target (Figure ‎3.24). Increases range from 5 basis points (i.e., 0.05% per annum) to 85 basis points (i.e., 0.85% per annum). In 42% of the cases, the increase amounts to 25 basis points. Only in the case of three non‑financial corporate issuances, the “penalty” for not reaching the target was a one-off payment of 10, 20, or 25, basis points. The money was to be donated either to eligible environmental organisations or local governments, or to be used to buy carbon credits and to fund green projects (and not paid to the bondholder as in all the other SLBs).

Interestingly, nearly half (42%) of SLBs in the sample had only one Sustainability Performance Target (SPT). Five corporate SLBs had three SPTs, and all the remaining bonds had two SPTs, including the two central governments SLBs in the sample. In Chile’s SLB, the penalty for not reaching each target was 12.5 basis points; in Uruguay’s SLB, there is a 15 basis points increase for not reaching each target (i.e., a penalty), but also a 15 basis points decrease for each target that is met (i.e., a premium). In the 26 SLBs included in the sample, the yield‑to‑maturity varied from 0% to 13.84% with the median at 5.36%.

Figure ‎3.24. Maximum coupon rate increase of sustainability-linked bonds

Source: SLB prospectuses, OECD analysis and calculations.

In nearly half (46%) of the corporate SLBs’ prospectuses in the sample and in two out of the four of official sector SLBs, a report on the performance against the KPIs relevant to the targets would be provided annually to investors. In the other corporate SLBs in the sample, such a report would be provided only after the targets were supposed to be met, or the prospectus did not mention a commitment to issue a report on performance. In the case of the two sovereign SLBs in the sample, the frequency of reporting is not annual for all KPIs because, as explained in the prospectuses, the assessments depend on costly data collection (e.g., satellite images). The annual disclosure of the performance is effectively a requirement by the ICMA Sustainability-Linked Bond Principles for evident good reasons. In some cases, it was also promised that an assurance provider would be hired to assure the quality of mentioned reports. While accounting and reporting on the issuer’s performance against relevant KPIs may be costly, in some cases the annual disclosure of this information – and not only when the target is supposed to be reached – may be material for investors. Particularly, this will be the case if targets are ambitious and the established penalties are relevant in relation to the yield of the issuance, because investors may incorporate the possibility of receiving a higher coupon in the future when pricing the SLB.

In any sovereign bond issuance, investors must be aware that it may be difficult for them to obtain or enforce judgements of foreign courts against a sovereign government. Specifically in the case of SLBs, it may even be challenging for investors to enforce the payment of the due coupon increase in the national judiciary of the sovereign issuer if a sustainability performance target has not been met. The existence of a bond contract may not be considered sufficient reason to limit the power of newly elected officials to legislate and to establish public policies that are different from the ones envisaged at the time of the bond issuance. Of course, a change in a SLB contract or the decision of not paying the coupon increase contractually due would probably harm the reputation of the sovereign issuer with investors, and this may be enough to deter any incompliance. Nevertheless, this legal and political uncertainty may explain why, as presented in Section ‎3.2, the issuance of SLBs among sovereigns is considerably less relevant than among corporate issuers.

3.3.5. The impact of sovereign sustainable bond issuance on the market

This section analyses whether the issuance of sovereign sustainable bonds has supported the development of a market of sustainable corporate bonds. It focuses mainly on green bonds, which is the most common type of bond issued by governments (OECD, 2023[35]), and on the non-financial corporate sector due to the fact that green bonds issued by financial companies have a different design: the proceeds are used for green loans instead of green projects. As a result, government green bonds are more comparable to those issued by non-financial corporations and, therefore, can arguably benefit more from government issuances.

Securities markets consist of a set of issuers requiring financing, investors looking to invest their savings and intermediaries that create a linkage between these investors and issuers. This occurs under a legal and regulatory environment that aims at facilitating price discovery, the secure ownership of holdings and the settlement of transactions. A developed market allows investors and issuers to trade with low transaction costs and assurances that minimise counterparty risk. The development of bond markets took years across OECD countries following the development of credible fiscal and monetary policy frameworks, the growth of government security markets, the design of sound legal and regulatory frameworks, and the evolution of the expertise of market intermediaries. While sustainable bonds are a relatively new instrument, they were developed in a period in which bond markets were already well developed, largely benefiting from the current market structure for conventional bonds.

Central governments, as typically the largest issuers of bonds in their respective jurisdictions, have had a notable market development role in bond markets. They aim at being a predictable and active provider of a large number of securities across the maturity spectrum, and at attracting a diversified base of investors to the domestic bond market. Additionally, these issuances allow market participants to use the yields paid by the government as a proxy for a benchmark yield curve of the “credit risk-free” interest rate. This curve is the main reference for the overall credit curve for non-sovereign bonds, greatly benefiting price discovery.

By issuing sustainable bonds, many governments aim to replicate their role in supporting the development of a conventional bond market to a sustainable bond market (refer to Figure ‎3.22). In the context of the sustainable bond market, such support may encompass the establishment of benchmarks for the framework, allocation, and impact reports, as well as raising awareness of sustainability issues.

The inception of a corporate sustainable bond market was largely neither driven by nor supported by central government issuances. In most cases, the opposite occurred – the government sector caught up with the private sector in the issuance of sustainable bonds. More specifically, of all the 79 countries that had at least one issuance of corporate sustainable bonds as of 2023, only in four countries (Hungary, Nigeria, Poland, and Togo) did the central government issue a sustainable bond before the private sector. Even when considering only green bonds and the non-financial sector, only in five countries did government issuance precede those from the non-financial corporate sector – namely in Belgium, Ireland, Hungary, Poland, and Nigeria.

Concerning the impact of government issuances on new corporate issuances, Figure ‎3.25 analyses trends in the growth rate of the amount issued in corporate green bonds four years before and after the first sovereign green bond issuance. Two observations can be made. First, only for the non-financial corporate sector is there a local peak in the amount issued in green bonds around the year of the first sovereign issuance. Second, there is no clear long-lasting effect of the issuance of sovereign green bonds on the growth of the issuance of corporate green bonds in either the corporate financial or the non‑financial corporate sector, as in the subsequent year of the sovereign issuance the growth rate reduces.

Figure ‎3.25. Median and range of the annual growth rate of the amount issued in green bonds

Note: This chart covers all countries with both corporate and sovereign green bond issuances. The trend line refers to the median growth rate of the amount issued in green bonds across these countries in the years before and after the issuance of the sovereign green bond issuance. The shaded area refers to the interquartile range – that is, the range between the 25% and the 75% percentiles of the growth rates.

Source: LSEG and Bloomberg, OECD calculations.

To complement the analysis above, Figure ‎3.26 compares the amount issued in green bonds as a share of all bond issuances by non-financial corporations across markets with and without sovereign green bond issuances. In the sample analysed, jurisdictions that issued sovereign green bonds before 2023 have always been in the top half portion of the chart. This means that the amount issued in green bonds as a share of all bonds issued by the corporate non‑financial sector has grown more in these jurisdictions after the year of their first sovereign issuance compared to most other jurisdictions that have never issued sovereign green bonds. Nevertheless, a few jurisdictions that have not issued sovereign green bonds did experience even stronger growth (e.g. Finland and Norway), suggesting that the issuance of sovereign green bonds can help but is not a necessary condition to boost the growth of issuances.

Figure ‎3.26. Increase in the ratio between the amount issued in green bonds to all bonds issued by non-financial corporations

Note: The base year for each chart is the year of the first issuance of a green bond by sovereigns displayed. This chart covers only a selection of 18 jurisdictions with a frequent issuance of non-financial corporate bonds to avoid outliers in the ratio of green to all bonds. Mexico issued its first sustainability bond in 2020, while Argentina and Brazil did in 2023. However, none of them issued green bonds.

Source: LSEG and Bloomberg, OECD calculations.

Lastly, Figure ‎3.27 contrasts the trend in the ratio between the amount issued in green bonds to all bonds issued by non-financial corporations for four sets of countries within the same continent. Overall, it shows that many jurisdictions did experience an increase in this ratio after the first green sovereign issuances, but this growth is by and large comparable to those experienced by countries without sovereign green bond issuances. For instance, this share increased in France more substantially only after 2020, three years after its sovereign issuance. In this same period (between 2017 and 2020), this share rose more in Italy and the United Kingdom, even though their governments only issued a sovereign bond in 2021. Similarly, in Sweden, the strongest rise in this share occurred between 2017 and 2019, prior to its debut sovereign issuance in 2020. A comparable increase was experienced by its neighbour Finland, but Finland has never issued a sovereign green bond. Similar trends of jurisdictions experiencing a large growth in their corporate sustainable bond market without sovereign issuances occurred in Asia and South America.

Figure ‎3.27. Trends in the ratio between the amount issued in green bonds to all bonds issued by non-financial corporations in selected countries

Note: Mexico issued its first sustainability bond in 2020, while Argentina and Brazil did in 2023. However, none of them issued green bonds.

Source: LSEG and Bloomberg, OECD calculations.

3.3.6. Liquidity

Although the sustainable bond market can benefit from the legal and regulatory environment for conventional bonds as well as the expertise of issuers and intermediaries, these elements can improve the development of the sustainable bond market only up to a certain point. Ultimately a market is only functional to the extent to which issuers and investors can find a counterpart to trade their security – that is, if markets are liquid. Market liquidity refers to the degree to which trading an asset impacts its value. The wider and deeper the issuer and investor base for an asset, the more likely it is for investors to buy and sell assets without meaningfully affecting the current price, which reduces the costs associated with entering or exiting a position in the market. Everything else held constant, there is a reinforcing loop in which the more issuers and investors are active in the market of an asset, the more attractive this market becomes for other investors and issuers due to the benefits of a liquid market. In that way, a diversified investor and issuer base is crucial for the development of markets.

This section assesses the liquidity of sustainable bond markets by comparing the bid-ask spread between a limited set of matched sustainable and conventional bonds for both the official and the corporate sectors. The set of bonds is the same as in the analysis in Section ‎3.3.2, and the matching controls for, among other variables, the issuer, currency, issue date, and maturity year. The bid-ask spread refers to the difference between the price at which participants offer to buy and sell a security, and it is the most used proxy for market liquidity and the one that conveys the most information in bond markets (Fleming, 2002[37]).

Figure ‎3.28 compares the weekly average bid-ask spread since issuance between sustainable and conventional bonds from the official sector. It shows that, out of the 19 matched bonds, in 14 of them the average bid-ask spread is higher for the sustainable bond compared to the matched conventional bond. This pattern was observed in multiple countries – namely France, Germany, Hungary, Lithuania, the Netherlands, Poland, Spain, Sweden, and the United Kingdom. Exceptions were observed in Poland and Serbia. Additionally, this trend also occurred over the maturity spectrum – it involves bonds maturing as early as 2025 and as late as 2051. The average difference was of 0.15 in the bid‑ask price, which represents roughly half of the average bid‑ask spread for the set of conventional bonds and 0.15% of the face value of 100 in the respective currency of denomination of the bond. It is worth noting that this difference in the average for the bid‑ask spread over time for paired bonds reflects a persistent tendency for the bid‑ask spread to be modestly higher for sustainable bonds (see Annex Figure ‎3.C.2, for bond‑specific time series data).

Figure ‎3.28. Bid-ask spread in matched official sector sustainable and conventional bonds

Note: The chart shows the difference in bid-ask spreads between bonds with 100 being the face value of the bond.

Source: LSEG and Bloomberg, OECD calculations.

For corporate bonds, there is also a small tendency for conventional bonds to have a lower bid‑ask spread compared to their sustainable bond pair. Figure ‎3.29 shows the distribution of the average differences in the bid‑ask spread between 214 pairs of sustainable and conventional corporate bonds. This set of pairs covers companies from various continents and sectors, and securities with varying maturities. Although the distribution of these differences is centred around zero, meaning that it is more likely to find a negligible difference in the bid-ask spread between the two types of instruments, a larger size of the distribution is on the right, which represents the cases in which the bid-ask spread is larger for sustainable bonds. A difference higher than 0.1 in the spread is nearly twice as likely to happen for sustainable bonds (22%) than for conventional ones (13%). On average the bid-ask spread is around 7% wider for sustainable bonds.

Figure ‎3.29. Bid-ask spread differences between matched sustainable and conventional corporate bonds

Source: LSEG and Bloomberg, OECD calculations.

Aside from differences in the investor base and the fact that there is a lower outstanding amount of sustainable bonds compared to conventional ones, one potential explanation for these results is that there are differences in the behaviour of climate-sensitive investors compared to conventional investors (Tang and Zhang, 2018[31]; Starks and Venkat, 2017[38]), with the former tending to have longer investment horizons than the latter. This means that climate-sensitive investors might be less willing to trade in times of asset depreciation to reduce their exposure, or in times of asset appreciation to profit from the short‑term movement. If this is indeed the case, it is more challenging to find a counterpart to trade in a market with climate-sensitive investors than otherwise, which can affect the liquidity of the securities traded by them.

The challenge in pricing some of the characteristics of sustainable bonds may also explain their lower liquidity. For instance, for investors to adequately assess the value of an SLB, they will need to understand the likelihood of whether a sustainability performance target will be met and, therefore, the present value of the possible coupon rate increase. While this is feasible, it represents a cost for the investor that, all else being equal, does not exist for investors in conventional bonds.

References

[5] Agence France Trésor (2023), Press Releases Archives, https://www.aft.gouv.fr/en/communiques-de-presse/archives.

[29] Ando, S. et al. (2022), Sovereign Climate Debt Instruments: An Overview of the Green and Catastrophe Bond Markets, IMF Staff Climate Note, https://www.imf.org/en/Publications/staff-climate-notes/Issues/2022/06/29/Sovereign-Climate-Debt-Instruments-An-Overview-of-the-Green-and-Catastrophe-Bond-Markets-518272.

[22] Associação de Investidores no Mercado de Capitais (2016), Brazilian Stewardship Code and Principles, https://amecbrasil.org.br/stewardship/amec-stewardship-code/?lang=en.

[41] Banca d’Italia (2021), Annual report on sustainable investments and climate-related risks, https://www.bancaditalia.it/pubblicazioni/rapporto-investimenti-sostenibili/index.html?com.dotmarketing.htmlpage.language=1 (accessed on 8 January 2024).

[40] Banca d’Italia (2021), The Responsible Investment Charter, https://www.bancaditalia.it/compiti/riserve-portafoglio-rischi/cis/index.html?com.dotmarketing.htmlpage.language=1&dotcache=refresh (accessed on 8 January 2024).

[24] Bolton, P. et al. (2022), Climate and Debt, Geneva reports on the world economy, Center for Economic and Policy Research, https://books.google.fr/books?id=ZU17zwEACAAJ.

[39] Bruckhaus, F. (2017), “SEC Issues Interpretations Relating to Rule 144A and Regulation S”, Oxford Business Law Blog, https://blogs.law.ox.ac.uk/business-law-blog/blog/2017/01/sec-issues-interpretations-relating-rule-144a-and-regulation-s.

[6] Bundesrepublik Deutschland (2023), Issuance Results in Syndication Process, https://www.deutsche-finanzagentur.de/en/federal-securities/issuances/syndicates.

[10] CBI (2023), Climate Bonds Standard Board, https://www.climatebonds.net/standard/governance/board.

[12] CBI (2023), Climate Bonds Standard V4.0, https://www.climatebonds.net/climate-bonds-standard-v4.

[34] CBI (2023), Reports, https://www.climatebonds.net/resources/reports.

[20] CBI (2022), Tax incentives for issuers and investors, https://www.climatebonds.net/policy/policy-areas/tax-incentives.

[14] CBI (2021), Climate Bonds Taxonomy, https://www.climatebonds.net/standard/taxonomy.

[17] Dikau, S. and V. Ulrich (2021), “Central bank mandates, sustainability objectives and the promotion of green finance”, Ecological Economics 184, https://doi.org/10.1016/j.ecolecon.2021.107022.

[32] Doronzo, R., V. Siracusa and S. Antonelli (2021), “Green Bonds: the Sovereign Issuers’ Perspective”, Banca d’Italia, MISP, Vol. 3, https://www.bancaditalia.it/media/notizia/green-bonds-the-sovereign-issuers-perspective/?com.dotmarketing.htmlpage.language=1&dotcache=refresh.

[30] Ehlers, T. and F. Packer (2017), “Green Bond Finance and Certification”, BIS, Bank of International Settlements, https://www.bis.org/publ/qtrpdf/r_qt1709h.htm.

[19] Elderson, F. and I. Schnabel (2023), “How green is our balance sheet?”, The ECB Blog, https://www.ecb.europa.eu/press/blog/date/2023/html/ecb.blog.230324~f417b719fd.en.html (accessed on 13 July 2023).

[33] ESMA (2023), “The European sustainable debt market – do issuers benefit from an ESG pricing effect?”, ESMA Report on Trends, Risks and Vulnerabilities Risk Analysis, https://doi.org/10.2856/468709.

[18] European Central Bank (2022), ECB climate agenda 2022, https://www.ecb.europa.eu/press/pr/date/2022/html/ecb.pr220704_annex~cb39c2dcbb.en.pdf.

[36] European Commission (2023), Regulation (EU) 2023/2631 of the European Parliament and of the Council on European Green Bonds and optional disclosures for bonds marketed as environmentally sustainable and for sustainability-linked bonds, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:32023R2631.

[23] Financial Reporting Council (2020), UK Stewardship Code, https://www.frc.org.uk/investors/uk-stewardship-code.

[21] Financial Services Agency (2020), Principles for Responsible Institutional Investors ≪Japan’s Stewardship Code≫ - To promote sustainable growth of companies through investment and dialogue, https://www.fsa.go.jp/en/refer/councils/stewardship/.

[37] Fleming, M. (2002), “Measuring Treasury Market Liquidity”, Economic Policy Review, Vol. 9/3, https://www.newyorkfed.org/medialibrary/media/research/epr/03v09n3/0309flempdf.pdf.

[26] Hinsche, I. (2021), “A Greenium for the Next Generation EU Green Bonds Analysis of a Potential Green Bond Premium and its Drivers”, Center for Financial Studies Working Paper 663, https://doi.org/10.2139/ssrn.3965664.

[3] ICMA (2023), Social Bond Principles, https://www.icmagroup.org/assets/documents/Sustainable-finance/2023-updates/Social-Bond-Principles-SBP-June-2023-220623.pdf.

[11] ICMA (2023), The Principles, Guidelines and Handbooks, https://www.icmagroup.org/sustainable-finance/the-principles-guidelines-and-handbooks/.

[1] ICMA (2022), Guidance Handbook, https://www.icmagroup.org/assets/GreenSocialSustainabilityDb/The-GBP-Guidance-Handbook-January-2022.pdf.

[2] ICMA (2021), Green Bond Principles, https://www.icmagroup.org/sustainable-finance/the-principles-guidelines-and-handbooks/green-bond-principles-gbp/.

[9] ICMA (2020), Governance Framework, https://www.icmagroup.org/sustainable-finance/membership-governance-and-working-groups/governance-framework/.

[15] IPSF (2023), The International Platform on Sustainable Finance – Annual Report 2023, https://finance.ec.europa.eu/system/files/2023-12/231204-ipsf-annual-report_en.pdf.

[7] Italian Treasury Department (2023), Risultati Aste e Collocamenti tramite Sindacato, https://www.dt.mef.gov.it/it/debito_pubblico/emissioni_titoli_di_stato_interni/risultati_aste/.

[28] Karpf, A. and A. Mandel (2018), “The changing value of the ’green’ label on the US municipal bond market”, Nature Climate Change, Vol. 2/161e165, p. 8, https://doi.org/10.1038/s41558-017-0062-0.

[25] MacAskill, S. et al. (2021), “Is there a green premium in the green bond market? Systematic literature review revealing premium determinants”, Journal of Cleaner Production, Vol. 280/124491, https://doi.org/10.1016/j.jclepro.2020.124491.

[35] OECD (2023), OECD Sovereign Borrowing Outlook 2023, OECD Publishing, Paris, https://doi.org/10.1787/09b4cfba-en.

[8] OECD (2022), “Climate Change and Corporate Governance,”, Corporate Governance, OECD Publishing, Paris, https://doi.org/10.1787/272d85c3-en.

[16] Sierra Club (2023), Lethal Investments: The Health Consequences of Cash Flows into Coal, https://www.sierraclub.org/sites/default/files/2023-08/SierraClub-CoalUtilityFinancingReport.pdf.

[38] Starks, L. and P. Venkat (2017), “Corporate ESG Profiles and Investor Horizons”, http://dx.doi.org/10.2139/ssrn.3049943.

[31] Tang, D. and Y. Zhang (2018), “Do shareholders benefit from green bonds?”, Journal of Corporate Finance, https://doi.org/10.1016/j.jcorpfin.2018.12.001.

[13] The People’s Bank of China (2021), Green Bond Endorsed Projects Catalogue, http://www.pbc.gov.cn/goutongjiaoliu/113456/113469/4342400/2021091617180089879.pdf.

[4] Ul Haq, I. and D. Doumbia (2022), “Structural Loopholes in Sustainability-Linked Bonds”, Policy Research Working Paper, World Bank Group, https://doi.org/10.1596/1813-9450-10200.

[27] Zerbib, O. (2019), “The effect of pro-environmental preferences on bond prices: evidence from green bonds”, Journal of Banking Finance, Vol. 98, p. 39e60, https://doi.org/10.1016/j.jbankfin.2018.10.012.

Annex 3.A. Methodology for data collection and classification

OECD Corporate Sustainability dataset

Sustainable bonds are mainly collected from LSEG. This dataset contains deal-level information of nearly 14 400 bonds issued by both the corporate and official sectors from 103 jurisdictions since 2013. This dataset provides a detailed set of information for each sustainable bond issuance, including the identity, nationality, and industry of the issuer; the type, interest rate structure, maturity date and rating category of the bond, the amount of and “use of proceeds” obtained from the issue. The issuance amounts were adjusted by 2023 US CPI.

For sustainable bonds, values for corporations, agencies, local governments, and multilateral institutions correspond to the “gross proceeds” (i.e., the amount paid by investors to acquire the bonds) in most cases. Where the information on the gross proceeds could not be retrieved, the “original amount issued” (i.e., the face value of the bonds in their legal documentation) has been used as follows:

i. for corporates, 22% of the amount issued from 2013 to 2023 corresponds to the original amount issued, whereas the remaining 78% corresponds to the gross proceeds. For that 78% in which the gross proceeds are used, the original amount issued is 2.9% higher. However, the amount issued in “all corporate bonds”, which includes conventional bonds, corresponds to the gross proceeds amounts in all cases.

ii. for agencies and local governments and multilateral institutions, 23% of the amount issued from 2013 to 2023 corresponds to the original amount issued, whereas the remaining 77% corresponds to the gross proceeds. For that 78% in which the gross proceeds are used, the original amount issued is 1.25% higher.

iii. for central governments, only the original amount issued (i.e., the face value of the bonds) is displayed for consistency with other OECD publications on sovereign debt markets.

Agencies and local governments include national government agencies (e.g., KFW, Transport for London), local governments (e.g., Prefecture of Shiga, Province of Ontario), and national development banks (e.g., China Development Bank, Brazilian Development Bank). This category does not contain US Municipals that are not readily available in LSEG. The category Central Governments includes governments, treasuries, and central banks. The category Multilateral Institutions includes organisations formed by three or more jurisdictions (e.g., Asian Development Bank, International Finance Corp), and the European Union.

The category “Asia excl. China and Japan” includes all jurisdictions in the continent excluding China and Japan (therefore this category includes Hong Kong (China), India; Korea, Singapore, and Chinese Taipei). “Latin America” includes jurisdictions both in Latin America and in the Caribbean. “Europe” includes all jurisdictions that are fully located in the region, including the United Kingdom and Switzerland but excluding Russia and Türkiye. “Other Advanced” includes all jurisdictions that are classified as advanced economies in IMF’s World Economic Outlook Database but that are not represented in the other categories in the figure (e.g., Australia, Canada, and Israel). “Others” includes mostly jurisdictions that are classified as emerging market and developing economies in IMF’s World Economic Outlook Database but that are not represented in the other categories in the figure (e.g., Saudi Arabia and South Africa).

LSEG data contains both Regulation S and Rule 144A sustainable bonds. Rule 144A presents a safe harbour from the registration requirements of the Securities Act for resales of securities not fungible with securities listed on a US securities exchange to qualified institutional buyers. Regulation S provides a safe harbour from the registration requirements of the Securities Act for offerings made outside the United States (Bruckhaus, 2017[39]). The calculations presented take account of this factor, and an exercise to eliminate the duplication when a single bond was issued both under Regulation S and Rule 144A was performed.

When calculating the outstanding amount of bonds in a given year, issues that are no longer outstanding due to being redeemed earlier than their maturity should also be deducted. Outstanding values refer to the “principal amount” or otherwise to the “original amount issued” (i.e., the face value of the bonds in their legal documentation) when the “principal amount” could not be retrieved. The early redemption data are obtained from LSEG and cover bonds that have been redeemed early due to being repaid via final default distribution, called, liquidated, put or repurchased. The early redemption data are merged with the primary bond market data via ISINs.

In Table ‎3.1, for the GSS bonds where more than one “use of proceeds” was disclosed by the issuer, the amount issued by the GSS bond was equally split for each of the use of proceeds. For example, if a GSS bonds amounting to USD 1 000 displayed clean transport and energy efficiency as promised use of proceeds, USD 500 was allocated into the category clean transport and USD 500 into energy efficiency one.

In Table ‎3.2, for the SLBs where more than one key performance indicator was disclosed by the issuer, the amount issued by the SLB was equally split for each of the key performance indicators. For example, if a SLB amounting to USD 1 000 displayed renewable energy and sustainable forest management as key performance indicators, half of that amount was allocated into the category renewable energy and half into the sustainable and forest management one.

In Figure ‎3.10, the values correspond to the sum of the amounts issued in 2022 and 2023. The five displayed categories stand for: Large listed, advanced (bond issuers that are MSCI World Index constituents as of September 2023), Large listed, emerging (bond issuers that are MSCI Emerging Markets Index constituents as of September 2023), Smaller listed (bond issuers that are listed on a stock exchange but that are not constituents of the MSCI World nor the MSCI EM indexes), and Unlisted subsidiary with a listed parent (bond issuers that were delisted prior to 2022, or that have never listed their equity, and which have their immediate or ultimate parent listed), Other unlisted company (other bond issuers that were delisted prior to 2022, or that have never listed their equity) . The bond issuer is categorised as listed if an ISIN exists and if it is associated with a valid LSEG Identification Code (RIC) made up of the bond issuer’s ticker symbol and an exchange code (based on the name of the stock exchange). If the bond issuer does not display an ISIN coupled with a RIC or its RIC shows a delisting year prior to 2022 (e.g., SPS.N^H97), it is classified as unlisted.

Sustainable investment funds

Open-ended fund and ETF data have been retrieved from Morningstar Direct. Funds retrieved as “Sustainable Funds” in the case their names contain climate or ESG relevant acronyms and words such as “ESG”, “sustainable”, “responsible”, “ethical”, “green”, “social”, “environment”, “Paris align” and “climate” (and their translation in other languages).

Funds without any asset value, as well as expired bond holdings, are excluded. USD (January 2024) is the portfolio currency for the holdings value.

Fixed Income funds are defined by Morningstar as open-ended funds and ETFs that have most of the holdings as fixed income assets.

A passive fund is defined by Morningstar as an open-end fund or exchange-traded fund whose portfolio managers do not choose specific investments but automatically invest in an index or part of a market. They often track stock or bond market indices.

Annex 3.B. An analysis of a premium for sustainable bonds

Annex Table ‎3.B.1. Paired GSS and conventional bonds for government primary issuances

	ISIN – GSS	ISIN – Conventional	Issuance – GSS	Issuance – Conventional	Maturity – GSS	Maturity – Conventional
France	FR0013234333	FR0013515806	2017	2020	2039	2040
France	FR0014002JM6	FR0013257524	2021	2017	2044	2048
France	FR001400AQH0	FR0013327491	2021	2018	2038	2036
Italy	IT0005542359	IT0005449969	2023	2021	2031	2031
Italy		IT0005436693		2021		2031
Italy	IT0005508590	IT0005496770	2022	2022	2035	2038
Italy	IT0005438004	IT0005530032	2021	2023	2045	2043
Germany	DE0001030757	DE0001102614	2023	2023	2053	2053
Germany	DE000BU3Z005	DE0001102515	2023	2020	2033	2035
Germany	DE0001030724	DE0001102614	2022	2022	2050	2053
Germany	DE0001030724	DE0001102572	2021	2022	2050	2052

Note: IT0005542359 is paired with a weighted average of IT0005449969 and IT0005436693.

Source: LSEG and Bloomberg.

Premium matching method

For this analysis, two datasets are used. One for corporate bonds and the other one for bonds issued by official sector entities. Both datasets include sustainable and conventional bonds, and the sample has been selected following (Bolton et al., 2022[24]), which considers whether the issuers had issued at least one sustainable bond and one conventional bond to ensure a balanced representation, to control for credit and exchange rate risk. To ensure data integrity and robustness in subsequent regression analysis stages, the focus is directed towards bonds with comprehensive data across specific variables. These variables include the issue date, yield‑to‑maturity (YTM), maturity date, coupon, coupon type and the amount issued in USD.

The Nearest Neighbour Matching technique was used for the matching process (using matchit command from R). This is a well-established method in propensity score matching analyses. A propensity score model was estimated using coupon, coupon class, issue year, maturity year and asset-linked securities type variables to estimate the probability of each bond being green. The matching was performed in a 1:1 ratio, meaning each green bond was matched to one conventional bond.

Propensity scores range between 0 and 1. In the context of propensity score matching, a propensity score is the probability of a unit (a bond, in this case) being assigned to a particular treatment given a set of observed characteristics. In other words, it is the conditional probability of receiving treatment (being a sustainable bond) as a function of observed characteristics. A score closer to 1 indicates a higher likelihood of being a green bond, given the observed features, while a score closer to 0 suggests a lower probability.

The objective of this technique is to achieve equilibrium in the distribution of observable characteristics between the treated group (sustainable bonds) and the control group (conventional bonds), thereby minimizing bias and enabling a more dependable comparison between the two groups. The aim is to simulate a randomized experiment, where the sole systematic difference between sustainable and conventional bonds is their "treatment" status, allowing any disparities in outcomes to be attributed to the “sustainable” status with a higher degree of conviction. The nearest method, also known as nearest neighbour matching, pairs a treated unit (sustainable bond) with a control unit (conventional bond) that has the closest propensity score.

Exact matching was executed on the two categorical variables, currency and issuer, to control for credit risk and exchange rate risk. This control ensured that the sustainable and conventional bonds matched were issued by the same entity and were denominated in the same currency, thus minimising potential sources of bias (Bolton et al., 2022[24]). The obtained dataset only includes bonds that have been matched based on their propensity scores and the exact match for issuer and currency, and thus have comparable characteristics, forming the basis for subsequent stages of analysis. The datasets include 72 government bonds (36 green and 36 conventional) and 1 736 corporate bonds (868 green and 868 conventional).

In summary, this methodology allows for a robust comparison between sustainable and conventional bonds, controlling for potential confounding variables. It ensures that possible differences observed in yield‑to‑maturity can be attributed mainly to the bonds' “sustainable” status.

Premium descriptive statistics

This section briefly presents the descriptive statistics of the data used for the regression performed following the methodology operated by Bolton et al. (2022[24]). The following descriptive statistics include 17 issuer’s domiciles for 17 different issuers and 7 different currencies for official bonds. For corporate bonds, descriptive statistics include 41 issuer’s domiciles for 419 different issuers and 24 different currencies.

Annex Table ‎3.B.2. Descriptive statistics – conventional matched bonds – official sector

	N	Mean	SD	Min	Max
Yield-to-maturity	36	1.8445	3.0878	-0.7472	13.3014
Coupon	36	1.6476	3.4468	0	7.1
Amount Issued in USD millions	36	3 308	3 346	26.36	10 192

Source: LSEG and Bloomberg, OECD calculations.

Annex Table ‎3.B.3. Descriptive statistics – sustainable matched bonds – official sector

	N	Mean	SD	Min	Max
Yield-to-maturity	36	2.0263	2.8278	-0.7601	13.5499
Coupon	36	2.9133	3.2413	0	7.29
Amount Issued in USD millions	36	3 935	3 969	27.70	14 592

Source: LSEG and Bloomberg, OECD calculations.

Annex Table ‎3.B.4. Descriptive statistics – conventional matched bonds – corporate sector

	N	Mean	SD	Min	Max
Yield-to-maturity	868	2.7310	2.0340	0.009	23.82
Coupon	868	2.7205	2.0309	0.01	23.82
Amount Issued in USD millions	868	238	447	0.948	5 293

Source: LSEG and Bloomberg, OECD calculations.

Annex Table ‎3.B.5. Descriptive statistics – sustainable matched bonds – corporate sector

	N	Mean	SD	Min	Max
Yield-to-maturity	868	2.7521	1.9617	0.001	17.52
Coupon	868	2.7361	1.9618	0.001	23.82
Amount Issued in USD millions	868	215	359	0.822	4 352

Source: LSEG and Bloomberg, OECD calculations.

Annex 3.C. The impact of sovereign sustainable bond issuance on the market

In Figure ‎3.26, the criteria are that a jurisdiction can only be selected if it had at least four years of green bond issuance by non-financial corporations. In addition, for issuers with sovereign green bond issuances, the country should have at minimum two issuances before and after the sovereign issuance with green bond issuances by non-financial corporations, which excluded countries whose first issuance was in 2021. This avoids the bias of using only one year to compute any ratio for the country. The ratio considers only the period between 2016 and 2023 and for years in this period with no green bond issuance but with conventional bond issuances by non-financial corporations, the ratio was set to zero.

Annex Figure ‎3.C.1. Bond-specific bid-ask spread time series for the official sector

Source: LSEG and Bloomberg, OECD calculations.

Annex Figure ‎3.C.2. Bond-specific bid-ask spread time series for the corporate sector

Note: Due to size constraints, only 72 bonds with a better matching statistic are displayed.

Source: LSEG and Bloomberg, OECD calculations.

Notes

← 1.  At the central government level, this reduction can be attributed to a scarcity of eligible expenditures for large issuers of sustainable bonds, such as Chile and France (OECD, 2023[35]).

← 2.  The sample encompasses both advanced economies and emerging market and developing economies for sovereign issuances, as well as corporations from varied sectors domiciled in multiple countries. The advanced economies for sovereign issuances include France, Germany, Italy, and the United Kingdom, while the emerging market and developing economies for sovereign issuances encompass Chile, Egypt, India, Mexico, and Türkiye. The corporate sector is represented by financial and non‑financial entities.

← 3.  Banca d'Italia provides an example of a central bank integrating financial and sustainability criteria into its investment portfolio strategy. The Bank publishes a Responsible Investment Charter, which establishes principles for merging financial and sustainability assessments (Banca d'Italia, 2021[40]), and the Annual Report on Sustainability Investments and Climate-Related Risks (Banca d'Italia, 2021[41]).

← 4.  Other jurisdictions with repo-eligible government sustainable bonds are Thailand (2 securities), Hungary (2) and Hong Kong (China) (1).

← 5.  Other jurisdictions with repo-eligible corporate sustainable bonds are Chinese Taipei (54 securities), Australia (8), Russia (5), New Zealand (4), Brazil (1) and the United States (1).

← 6.  Other jurisdictions with repo-eligible agencies and supranational sustainable bonds are Australia (11 securities), New Zealand (7), Thailand (6), Indonesia (4), Russia (3), Chinese Taipei (3), the United States (1) and Viet Nam (1).