The two most broadly used classification schemes to analyze economics issues of a set of countries are geographical classification and income classification. Geographical classification has been used under the assumption that spatial location is interlinked to social and institutional factors which affect economic performance. Therefore, economies are not isolated from a spatial dimension. This classification is useful to explain different stages of economic progress by considering a country’s industrial development. Income classification is a practical method to cluster countries, since it helps to better understand economics issues in a context where economic wealth is relatively similar.
However, these types of classifications are not the most accurate to analyze sustainable development because the latter considers social, institutional and environmental factors in tandem.
Why do we care about sustainable development if we assess sovereign risk? The main reason is that there is a global and, particularly, a financial market recognition of the importance to include environmental (E), social (S), and governance (G) performance into sovereign risk analysis. For that reason, we are investigating the impact of ESG factors in the very long-term sovereign risk. The rationale is that social and governance performance give valuable insight into the willingness to pay, and environmental risk could affect both country’s willingness and capacity to pay its debt and interest on time and in full.
This insight shows the first step of this research by performing an innovative country classification which considers only ESG factors. We use three indicators to assess the ESG performance. These indicators are composed of 48 quantitative variables defined as scores scaled from 0 to 100 , indicating the economy’s performance vis-à-vis specific risks:
- (E) performance is composed of three dimensions: energy policy, climate risk and natural resources. Energy policy considers energy as a production factor that has direct and indirect effects on economies and societies. It captures the government’s efforts in terms of access to affordable energy and use of renewable energies. In the long-term, it measures the inclusivity and sustainability orientation of the energy policy. Climate risk assesses the exposure of countries to climate change and measures the financial materiality of these risks. According to the TCFD climate-related risks are divided into risks related to the physical impacts of climate change, and risks related to the transition to a lower-carbon economy. Natural resource endowment provides information about essential services, notably the provision of food, fibers, clean air and purified water.
- (S) performance includes five dimensions: Human capital & innovation measure the country’s capacity to generate new technologies and high value-added production. Health measures the country’s capacity to maintain the health of its population. Inequality considers distortion of income distribution and resource allocation. Societal evaluates the country’s progress in terms of the society’s freedom. Finally, employment measures the country’s capacity to provide a job for the entire working population, and then maximizing the potential output.
- (G) risk includes control of corruption, government effectiveness, rule of law, regulatory quality, political stability & absence of violence, and voice & accountability. Evaluation of these aspects allow us to identify to what extent the relationship between the government and the society is transparent and efficient, how agents perceive the rules of society and to what extent citizens exercise their freedom of expression.
The database is composed of 146 countries in a quarterly frequency from the fourth quarter of 1999 until the second quarter of 2018.
The methodology used to classify countries, according to their environmental, social and governance performance is the Hierarchical Ascendant Classification (HAC). This methodology is useful to identify clusters in a database. In our case, observations x_it are scores that belong to a country i in a period t. Distance is measured according to the Ward Methodology which minimizes the distance between two clusters in each iteration. We apply K-means methodology to allocate each observation to the cluster with the nearest average.
The main conclusion is that countries are differentiated mostly by its governance risk and social performance that help us to characterize them in: (i) Highly Institutionalised, (ii) Politically Stable with Social Challenges (iii) Fragile Governance, and (iv) Fragile Governance and Not-Inclusive countries (See Map below).
ESG Country Classification (for 146 countries)
Note: Weighted average was used to have a global classification.
The methodology allows us to follow the trajectory of each country because it is a dynamic classification. For instance, Brazil has changed from Politically Stable with Social Challenge to Fragile Governance country group since 2014. This is explained by corruption scandals related to the government and the political instability that deteriorated the governance performance. The same trajectory is given for Argentina, in which the 2001 currency crisis fragilized the governance performance of the country until now. Another interesting case study is Greece. The crises that started in 2009 have negatively affected the social performance in the country which passed from Highly Institutionalized to Politically Stable with Social Challenges country group.
This classification gives valuable information about a country performance regarding its ESG behavior. By combining ESG factors with macro financial performance, we have a better comprehension about country’s capabilities to face to shocks in the short- to long-term.
Gabriela Aguilera and Julien Moussavi, Department of Economic Research
Source: Aguilera G. and Moussavi J., 2018, “Are ESG factors structural determinants of economies?”, Beyond Ratings Research Note, September.