Think differently

ESG & sovereign bonds: a new paradigm?

Lionel Martellini , Professor, EDHEC-Risk Institute former CEO

Over the past ten years, sustainable and responsible investment (also called ESG, for its Environmental, Social and Governance criteria) has become unavoidable and is gaining increasing favour…

Reading time :
2 Jun 2022

Over the past ten years, sustainable and responsible investment (also called ESG, for its Environmental, Social and Governance criteria) has become unavoidable and is gaining increasing favour among investors. The share of global ESG holdings in mutual funds and EFTs (1) doubled between 2015 and 2020 (2), reaching a total asset value of nearly 1.7 billion dollars by the end of 2020 (3). By comparison, at the end of that year, the total value of all managed funds in the world was 103.1 billion dollars (4). This signals awareness of the crucial role that the finance sector can play in the transition to a low-carbon economy that tackles global warming.

Understanding the ESG criteria

Nevertheless, while ESG criteria have been integrated into share and corporate bond markets, they are not yet systematically taken into account in risk analysis and investment decisions for government bonds. The reasons include a lack of understanding among investors of how to integrate ESG criteria into the analysis of national debt on the one hand, and a lack of coherence in the definition and measurement of relevant ESG criteria on the other.

In a new study launched by the EDHEC Risk Institute (5) and published in 2021 in the Journal of Portfolio Management (6), Lionel Martellini, Professor of Finance at EDHEC and director of the EDHEC Risk Institute, and Lou-Salomé Vallée, Ph.D. in Finance (EDHEC), analysed the integration of ESG criteria into risk management and investment decisions involved in sovereign bonds.

The main purposes of this research are, first, to assess the impact of ESG criteria on sovereign bond risk and return indicators from the investor’s point of view, and second, to assess the possibility of integrating ESG restrictions into sovereign bond portfolios with the aim of improving ESG scores without significantly increasing either the absolute (volatility) or relative (tracking error) (7) risk budgets, or substantially reducing expected returns.

Deeper analysis of a decade of ESG investments

This study investigates 35 countries - 20 developed and 15 emerging economies – over a ten-year period (2010-2020). The ESG data (8) are used to evaluate the Environmental (E), Social (S) and Governance (G) criteria of the states involved. Each E, S and G score corresponds to a weighted sum of underlying indicators. Thus, the E score assesses the capacity of a country to respect the environment and to measure its natural resources using indicators such as water and waste management. The S score measures a country’s respect for human rights using indicators such as freedom of opinion and expression, and the rights of minorities. Finally, the G score assesses a country’s enactment of safety and regulation, using indicators such as the rate of corruption and the efficacity of the legal system. Each score is a grade between 0 and 10 (the best mark).

From their analysis of these data, the authors draw four main conclusions.

First, the research results show that there is a negative relationship between a country’s E, S and G scores and the return on bonds for an investor who buys a bond and either keeps it until its maturity (the long-term perspective) or hopes to resell before then (short-term perspective). Put another way, the higher a country’s E, S and G scores (i.e. the less it is exposed to environmental, social and governance risks), the lower the return on the country’s bonds, regardless of the investment time frame. This relationship is confirmed for the E and G scores of developed economies, and for the S scores for emerging economies. These results are consistent with the return/risk relationship in which the trade-off for any financial gain is acceptance of a certain degree of risk or uncertainty that any such gain will be made. Thus the E, S and G criteria apparently help to explain the differences among national issuers.

In line with these results, the authors came to three other conclusions about sovereign bond portfolios.

On the one hand, the E, S and G criteria in sovereign bond portfolio strategies create an opportunity cost. To improve the ESG profile of a portfolio, different tactics can be used as positive or negative screening methods (9) or portfolio optimisation methods (10).

One negative ESG screening strategy (in which the investor totally excludes investment in bonds issued by the 25% of countries with the lowest ESG scores), leads to more diverse portfolios, lower tracking error results (7), and a slight improvement in the ESG profile of the portfolios. One positive ESG screening strategy (whereby the investor invests only in bonds issued by the 25% of countries with the best ESG scores), leads to a greater improvement in the ESG profile of the portfolios, at the price of an increase in the absolute and relative risk of the portfolios because the investment universe is tighter.

Moreover, the ESG scores of sovereign bond portfolios can be improved by adding ESG score restraints to a portfolio optimisation strategy. The study shows that for the same improvement in ESG scores (especially for S and G components), the most effective of the screening methods is a so-called Minimum Variance Portfolio strategy, which consists of obtaining an efficient portfolio with a minimum “volatility” risk.

Finally, to reduce further the opportunity costs mentioned, so-called ESG momentum strategies (11) can also be used in the sovereign bonds market.

In total, limited cost and risk

This study demonstrates that taking ESG criteria into account in sovereign bond investment involves an opportunity cost.  It confirms that « doing good and doing well », are generally not the same thing. Therefore, allowing investors to integrate ESG criteria into their investment decisions without unreasonable performance cost has become crucial.

The research was made possible by the support of Amundi, under the rubric of the EDHEC Risk Institute’s research chair on “ETF, Indexing and Smart Beta Investment Strategies”.

(1) Exchange Traded Funds: investments funds comprised of stocks and bonds issued by dozens of firms and or states.

(2) Broadridge, ESG: Transforming asset management and fund distribution, September 2020.

(3), ESG funds defy havoc to ratchet huge inflows, 6 February 2021.

(4), Global asset management industry report: The $100 Trillion Machine, 8 July 2021.

(5) EDHEC-Risk Institute Publication, March 2021

(6) Lionel Martellini (EDHEC) and Lou-Salomé Vallée (EDHEC), The Journal of Portfolio Management Novel Risks 2021, jpm.2021.1.290; DOI:

(7) A tracking error is a measurement of relative risk used in the management of share portfolios that are indexed or compared with a benchmark portfolio. It represents the annualised standard deviation of a portfolio from the benchmark. The higher the tracking error, the further away the portfolio’s average performance from the benchmark. Conversely, a low tracking error shows that the portfolio’s performance is more or less equal to that of the benchmark.

(8) Verisk Maplecroft (provider of the ESG data)

(9) Screening methods are investment filtering methods. In the context of this research, they allow countries to be selected on the basis of extra-financial criteria (the ESG criteria), with the aim of creating a specific investment universe. A negative screening is an exclusion practice (the exclusion of the 25% of countries with the lowest ESG scores), and a positive screening is a selection practice (the selection of the 25% of countries with the highest ESG scores).

(10) Portfolio optimisation is the process of selecting the best portfolio (asset allocation), among all the possible portfolios, according to the goal. The aim generally is to maximise factors such as expected return and to minimise costs such as financial risk.

(11) ESG momentum is the percentage of improvement or deterioration in the ESG score of a country in a given period (in this case, in the preceding twelve months). In this study, an ESG momentum strategy consists of buying bonds issued by the 15% of countries with the best ESG scores and to sell bonds issued by the 15% of countries with the worst ESG scores.

Other items you may be
interested in


The fine print on interest-free loans

  • Tristan-Pierre Maury , Professor

Predictive justice: failsafe or false promise?

  • Christophe Roquilly , Professor, Honorary Dean of Faculty, Director of the EDHEC Augmented Law Institute