Fitch takes ESG to a new level

A holistic approach to integrating environmental, social and governance (ESG) factors into credit processes is becoming the norm across the global debt market. Fitch Ratings (Fitch)’s integrated scoring system has been operational for close to six months and Andrew Steel, the rating agency’s London-based managing director and global head of sustainable finance, explains why the agency believes this system could transform ESG investing.

You recently visited Australia for client meetings and to be part of Fitch’s annual Australian event, Fitch on Australia. Presumably getting an update on ESG in capital markets was one of the aims of the trip, too. What are the most pressing ESG issues from Australia that you are taking away with you?

My overall take-away is that interest in Australia is still at an early stage. I was a little surprised by this, given Australia’s relatively heavy reliance on the mining and natural-resources industries.

When I talk to investors around the world, I get the sense that there is more urgency on ESG in other countries than there is in Australia. This could be to do with reporting, but also perhaps the lower proportion of international investors participating in the Australian domestic market. Many large international investors are under pressure to report on ESG to their asset owners. While there is some evidence of this in Australia, I think it is definitely at an earlier stage than in many other developed markets.

Many of the conversations that are happening in Australia are on sustainable loans and green bonds, while the broader global trend of ESG investing is towards integration by asset managers. In some other parts of the world, ESG investing is becoming embedded in investor processes and I expect Australia to follow.

Fitch integrated ESG analysis across its general rating process from the start of 2019. What is the rating agency’s role in market development?

The integration was a big piece of work, the catalyst for which came from the UN Principles for Responsible Investment’s ESG in credit risk and ratings initiative. This aimed to enhance the transparent and systematic integration of ESG factors in credit-risk analysis. The message that emerged loud and clear was that rating agencies are not displaying subcategories of risk sufficiently clearly.

We were being asked to provide three sets of information: which aspects of ESG risks are relevant from a credit perspective, how material they are to decision-making and whether there is systemic risk for a sector.

We spent several months in consultation with asset managers globally to try to ascertain whether they would like to have this information on a permanent and ongoing basis. We very quickly reached the conclusion that this was not a fad, rather it forms part of asset managers’ requirements and will continue to do so going forward.

We decided fully to integrate the analysis across our whole suite of credit research in a system which extracts the elements of ESG risk from our existing criteria and displays them in a separate template. This has around 15 general ESG-risk issue categories.

These categories include greenhouse-gas emissions and air quality, water and wastewater management, environmental-catastrophe risks, employee wellbeing, affordability, social impacts, governance structure, management strategy, and financial transparency.

Were these risks previously considered in rating analysis but are now being broken out as separate and specific information?

That’s correct. Our board considered carefully whether this would provoke negative feedback from issuers. We aren’t giving issuers a choice – if they take a rating they will automatically receive this extra layer of analysis. While they can appeal the overall credit rating they won’t have an opportunity to appeal the ESG scores separately.

An important point is that any factor we identify as a driver or key driver within credit analysis was already being considered. None of what is being produced has resulted in a change to existing ratings.

How many scores have you published under the system so far and how has the scoring system been working?

We have created more than 70 assessment templates and published ESG-relevance scores for entities rated by all our analytical groups – apart from structured finance, which is still in development. We have published more than 73,000 individual ESG-relevant scores on more than 5,300 publicly rated entities since January this year.

The ESG scores we produce are on a five-point scale. The highest score – five – indicates the presence of an individual ESG risk factor which by itself has caused a different rating level to be assigned. We call this a key rating driver.

Level four relates to aspects of risk which by themselves are not sufficient to change a rating but may change the rating in conjunction with other aspects. It could be that several of these risks together cause a change to the rating or that one of these combined with a traditional risk – such as an extremely weak financial profile – is enough to trigger a change.

Our score of three occurs most frequently. This covers risks that are either minimally relevant to ratings or which are being effectively managed by entities to avoid any impact on the credit profile.

Below the score of three are two further levels: two, where the risk element is not relevant to the entity but may be relevant to the sector, and one, where the ESG risk is not applicable from a credit perspective even on a sector basis.

“Many large international investors are under pressure to report on ESG to their asset owners. While there is some evidence of this in Australia, I think it is definitely at an earlier stage than many other developed markets.”

What time horizon is applied to risks in your analysis?

There is a vast array of ESG rating products from more than 220 providers, which are based on a range of very different time frames. This can be confusing for asset managers as they try to work out the relative importance of different factors. To provide consistency, we decided to match our scores with our rating time frame – typically a 3-5 year forecast period.

What outcomes has the analysis produced so far?

The analysis shows some interesting and perhaps surprising trends across our ratings. For instance, more credit ratings for natural-resources companies globally are being affected by governance and social factors than environmental ones. This is because these companies typically know the financial consequences for mismanagement of environmental regulation and thus manage it actively. Credit problems tend therefore to occur only where institutions are thrown a curveball through events they weren’t anticipating.

Can you give some insight into how the scores you are producing might develop over time?

The example of the impact of water and wastewater management on global airlines is a good one to answer this question. From a credit perspective, these issues are not currently relevant to the sector or its entities, so all airlines are currently assigned a score of one. Airlines use water and produce wastewater but not to an extent that it could conceivably affect the credit profile.

However, the score would be a two if airlines started to face new regulation around the level of wastewater they emit into the atmosphere. In other words, the issue would be relevant to the sector but it is not yet having an impact on any entities.

If detailed regulations were then published which included a significant financial penalty for breaches, airline entity scores would disperse between three, four and five.

Airlines that were well prepared to demonstrate compliance would get a score of three: they are managing the risk in a way that will not affect their credit profile.

Moving down the spectrum, we might assign a score of four to entities that had to commit capex to demonstrate compliance with the new regulations if, for example, the capex reduced the rating headroom but didn’t cause a rating change.

We would probably score a five for entitles likely to be fined or compelled to make changes to operating procedures, resulting in a negative financial impact and change of rating.

Is the system designed to be positive as well as negative?

Positive scores can occur but they are rare. A good example is UK regulated water utilities, where the regulator outlines a pricing formula mechanism. A main component of this is reduction in water leakages. If companies outperform they can receive additional revenue, which could generate a positive credit score under the water-management criteria.

How are second- and third-order impacts treated?

The scoring system doesn’t account for these and it was not designed to. It focuses on potential credit impacts on specific entities and an entity’s ability to manage those impacts. Factors that do not have a direct impact fall to another entity to manage. It may be that the entity we are reviewing ends up contributing to an issue but it is not considered relevant if it does not have an impact on the credit profile.

It may be that an entity is profitable and benefits from a good credit profile despite a poor ESG environment. Take for example a ride-sharing app that ruthlessly exploits its workforce in an emerging-market country. If there is no perceived credit impact in the short-to-medium term, it potentially won’t have a poor ESG relevance score from a credit perspective even if it is engaged in negative social practices.

The scores we produce are concurrent with our ratings and are maintained by our credit analysts. As longer-term issues for a sector start to have a credit impact our analysts will reflect these in the ESG scores as and when they are factored into our forecasts. This is important for debt investors that are primarily concerned about mitigating downside risks which could have a negative impact on credit profiles and thereby affect debt-repayment capacity.

“To be able to isolate the ESG credit characteristics of a specific set of worldwide entities is very powerful. We will be looking to identify what it is about entities that sets them apart from their peer group.”

How do the scores Fitch assigns compare between developed and emerging markets, including the relative prominence of the individual components of ESG?

Overall, 22 per cent of corporate entities have at least one score of four or five, which breaks down into 19.3 per cent with scores of four and 2.7 per cent with at least one five. This confirms that, from a credit perspective, ESG risk factors are important and material to credit ratings.

Governance dominates in developed-market Asia and the Americas, closely followed by social-risk impacts and with environmental risks much less prominent. This is perhaps not surprising as environmental risks are often heavily legislated for in developed markets and so are more visible and manageable.

Social-risk impacts are slightly greater than governance in developed-market Europe. We think this is probably related to the high level of public awareness in Europe combined with populist governments that react to media sentiment. Sudden changes in legislation can occur as a result of public concerns – for example, a focus on obesity that has resulted in a sugar tax for food manufacturers.

This type of impact is difficult for companies to predict or plan for, so it can have a greater negative effect on credit profiles. Additionally, there is a higher incidence of strikes, boycotts and rapid changes of consumer sentiment in some sectors across Europe.

Governance is the central concern in emerging markets. It is not that legislation doesn’t exist in emerging markets, but rather that fines for environmental transgressions tend to be smaller relative to the size of the company.

What comes next for Fitch’s ESG offering now the rating framework is operational?

What’s next is very exciting as the scores we are creating will start to become a consistent time series of data, which we can drill down into, looking for patterns and correlations between ESG and credit changes.

To be able to isolate the ESG credit characteristics of a specific set of worldwide entities is very powerful. We will look to identify what it is about entities that pick up a score of four or five that sets them apart from their peer group. Perhaps it is due to individual operational practices, governance standards or financial-risk practices. 

We will also be able to look more deeply at individual aspects of ESG risk – to determine where the biggest credit impacts are globally and whether these can help us predict where they might start to occur next.

From what Fitch has seen, what are the hidden risks that investors may not have assessed or priced when it comes to performance impact across their portfolios?

I would argue that in many cases it is the other way around. Large global asset managers with fully integrated ESG processes are often frustrated because pricing tends to under- or overshoot. They argue that lack of consistency and clarity of information can cause aspects to be priced in that are not relevant.

A full integration strategy may perform worse in a bull market and better in a bear market. This lends itself to the idea that pricing is being skewed by lack of clarity around information on credit relative to overall sentiment on an entity.

Investors tell us that integrated ESG scoring on an entity-by-entity and sector-by-sector basis is very helpful because it enables them easily to distinguish credit impacts. If two credits look the same but price differently this data can potentially help pinpoint where and why. It could be an entity-specific risk not previously identified or that the sector is being priced up based on sentiment risk.

To have access to consistent, granular data on credit globally should help move the debate forward by helping to identify some of these risks.