15 minute read
We look at how and why 2019’s Amazon rainforest fires and protests in Chile led to underperformance of sovereign assets.
ESG policy lessons from Latin America
Environment, Social and Governance (ESG) issues are not only increasingly impacting corporations and their funding costs, but sovereigns as well. In this paper, we take advantage of two important events from the Latin America region (LatAm) in 2019 as case studies to illustrate the multi-faceted effect that the rise of ESG is having around the world. Both are very interesting episodes given the role that environmental and social factors play in each economy:
Case study 1: Fires in the Amazon rainforest
These widespread fires highlighted the rapid increase in the rate of deforestation in the Brazilian Amazon in late August 2019. We’ve selected this as our first case study because the Amazon rainforest has critical ties that link the health of rainforests to the rest of the world, as many studies have shown.1
Case study 2: Social protests and violence in Chile
The Chilean social protests exploded in the middle of October 2019 and ultimately led to violence across the country. It’s a useful case study because in Chile the historically stable democratic system and growing middle classes have been key factors behind its solid macro performance low borrowing costs compared to its emerging market (EM) peers.
How did these events impact their country’s ESG credentials?
The events of both cases challenged the environmental and social market assessments of both countries. But did investors care? More directly, is there evidence of underperformance of these countries’ assets as the respective situations deteriorated?
For the answer, we’ve looked at the performance of Brazilian and Chilean dollar-denominated spreads versus its peers (to control for any other factors) during the weeks and months following the events: August and October of last year in Brazil and Chile, respectively. In this paper we also provide some anecdotal evidence of the impact of these events on the holdings of the countries’ sovereign bonds by large institutional investors.
The events of both cases challenged the environmental and social market assessments of both countries. But did investors care?
Relative underperformance of spreads
The evidence indicates that there was spread widening/underperformance versus peers in the immediate aftermath of these events:
- Brazilian bonds underperformed Mexico by around 40 basis points (bps) following the news of widespread fires in the Amazon
- Chilean bonds underperformed Peru by around 20bps following the escalation of domestic violence
Hard currency borrowing costs
These changes materially increased the hard currency borrowing costs in each country, and there is evidence of reductions in the holdings by large institutional investors. These direct financial effects of ESG factors on credit spreads of companies and countries are a specific focus of our research and we will examine the impact of corporates in upcoming publications.
ESG case study on the ‘E’ angle: Fires in the Amazon rainforest
Key changes to Brazil’s environmental policies began quickly under Bolsonaro
On 1 January 2019, President Bolsonaro took office and began to make key changes to Brazil’s environmental policies. One policy change was the halt of demarcated lands for indigenous reserves:
Bolsonaro transferred responsibility for delineating indigenous territories from the Justice Ministry to the Agriculture Ministry, which one lawmaker described as “letting the fox take over the chicken coop".
The decree also transferred the agency for indigenous affairs, known as FUNAI (Fundação Nacional do Índio), from the Justice Ministry to a newly created Ministry for Family, Women and Human Rights headed by an ultraconservative evangelical pastor. FUNAI oversees health care, housing and language preservation for indigenous groups.
This policy also caused climate groups to call Brazil’s commitment to the Paris Climate Accord insufficient, as plans to open up the Amazon for greater development would likely make it impossible for Brazil to meet its reduced emissions pledge. In addition, Bolsonaro filled key cabinet positions with officials openly hostile towards the fight against global warming. Lastly, Brazil cancelled its offer to host the 2019 Climate Change Summit just before the new president took office.
News of widespread fires caught global attention
On 18 July, Brazil’s National Space Research Institute (INPE) reported that preliminary satellite data released showed that more than 1,000 square kilometres (400 sq. miles) of the rainforest had been cleared in the first 15 days of July, representing a 68% increase from the entire month of July 2018. In response, President Bolsonaro accused Brazil’s INPE of lying about the scale of deforestation3 and then fired the director of Brazil’s INPE on 2 August.4
Importantly, around mid-August, news of widespread fires in the Amazon started to come to international attention. Specifically, NASA’s Biospheric Sciences Laboratory confirmed the previous results by its Brazilian counterpart of:
…A noticeable increase in large, intense, and persistent fires burning along major roads in the central Brazilian Amazon [which] are more consistent with land clearing than with regional drought.
This news in turn highlighted the rapid rate of deforestation of the Brazilian Amazon during the Bolsonaro administration, which had grown by 30% from 2018 to almost 10 thousand square km, as reported by Reuters.6 By the end of August, President Bolsonaro announced that he was mobilising the army to help fight Amazon forest fires after growing international concerns.7 By then, the damage to Brazil’s environmental reputation had suffered significantly.
Market impact: Investors starting noticing…
Drawing relative comparisons between Brazilian and Mexican spreads Between 1 August and 15 October 2019, when the news surrounding the Amazonian fires became more public, Brazilian dollar bond spreads increased by around 5bps at the front-end of the curve and 20bps in the long-end – i.e., the curve bear steepened (Chart 1). Importantly, during the same time period Mexican spreads, which we think are a good comparison, actually tightened by 20bps at the front-end and 15bps at the long-end (Chart 2).
Clear underperformance of Brazilian bonds versus Mexican bonds
We think this type of relative comparisons is a key component for ESG analysis on credit, as in some cases the overall trend only tells part of the story. In this case, there was a clear under-performance of Brazilian bonds versus Mexico from ~20bps to ~60bps at the front-end, and to an even greater degree in the long-end, from ~25bps to ~85bps (Charts 3 and 4).
While causality is difficult to prove, Brazilian bond under-performance tightly followed the news of the Amazon fires
As is typical with analysing ESG factors, it is often difficult to separate the influence of these factors from other events. In this case, there is likely some influence on Brazilian spreads to headlines surrounding the approval of the pension reform. However, the dates correspond materially to the headlines around the Amazon – specifically, there is an uptick in the differentials right on 22 August when NASA confirmed the results, so again we feel the evidence is compelling.
Lastly, we have observed that the negative impact on Brazil’s bonds seems to have been temporary, especially at the front-end, where spreads declined somewhat after November. This could be because while international pressure did force Bolsanaro to change policies at the end of August,8 and it took some time for investors to regain faith in the change and for the changes to be implemented and evidenced.
Backlash from institutional investors
Away from market indicators, the Amazonian fires in Brazil prompted a strong backlash from institutional investors. For example, the 230 member network of the Principles for Responsible Investment with over $16 trillion in assets under management (AUM), issued a statement warning companies “to either meet their commodities supply chain deforestation commitments or risk economic consequences”.9
Backlash from sovereigns
At the sovereign level, Reuters reported Nordea Asset Management suspended on its purchases of Brazilian government debt in response to the fires on 30 August.10 Nordea joined other Nordic investors in taking action such as KLP, a Norwegian pension fund with over $80 billion in assets under management, and Norwegian insurer Storebrand. At the national level, Germany and Norway together froze roughly $70 million in planned aid assistance to Brazil.
Backlash from large corporates
Finally, many European companies re-examined their supply chains for Brazilian leather goods and other products.11 Even in the US, there was a reaction. For example, the Guardian reported that “US clothing company VF Corporation, which is behind brands including Timberland, Kipling bags and The North Face, has suspended Brazilian leather purchases” in response to the fires.12
Combined, these behaviours from both investors and companies support the empirical evidence of Brazil’s underperformance as shown above, and are increasingly clear evidence of the role of ESG in investment decisions.
ESG case study on the ‘S and G’ angles: Social protests and violence in Chile
Protest over transportation fares quickly escalated
What started as a small group of students protesting a hike in transportation fees in Santiago on 18 October 2019 quickly escalated into damages to metro stations and violent confrontations with the police. By that weekend, the government imposed curfews and took more aggressive action against a growing number of protesters. Rioting quickly began spreading to smaller cities mostly by youth coordinating through social media. Clearly the roots of the social discontent were deep and tension quickly erupted.
Daily marches across the country with various degrees of violence and property damage took place over the following weeks, leading to over one million people taking over the streets to demand the President’s resignation on 25 October.
Growth and the currency value were negatively affected
All together the violence took a heavy toll as 30 people died, almost 2,500 had been injured and economic activity quickly deteriorated. In fact, GDP contracted by 3.5% year over year in October and November, dragging growth in the fourth quarter to -4% versus the previous one, the worst print in over a decade. The Chilean peso weakened by more than 16% by the end of November, after which the central bank implemented a historic FX intervention program to control against further depreciation.
The government made important concessions
The government of President Pinera was slow to respond at first, but increasingly gave concessions to the protesters. Key Cabinet posts, including the Minister of Finance, were replaced and significant increases in health and social expenditure were approved (see our take here). This involved larger fiscal expenditures and debt burden. Finally, on 15 November, the National Congress signed an agreement to hold a national referendum in April 2020 regarding the creation of a new Constitution.
Market impact: A big surprise for investors…
The scale and violent turn of the social protests in Chile took most investors by surprise. The surprise was all the greater since Chile is regarded as a stable democracy, where social pressures are thought to be addressed through participatory processes. In fact, recent elections have seen smooth power transitions between left and right leaning administrations, both focused on orthodox macro policy-making.
Like in our first case study, there was a negative reaction on Chile’s spreads, and the correlation to the events is even clearer than the in Brazilian example.
With hindsight, we can see that as the country’s income expanded, the gap between expectations and the reality of the average Chilean’s prosperity had been widening. We think that the events reflected social and governance issues that had been largely dismissed by the market. By itself, that is an important lesson for longer-term sovereign investors.
Chilean assets rapidly adjusted
Like in our first case study, there was a negative reaction on Chile’s spreads, and the correlation to the events is even clearer than the in Brazilian example. In particular, between 15 October when the first protests started to emerge on the headlines and December, Chilean dollar spreads, which were very low to start with, widened by around 10bps in the most liquid bonds (from the mid 20bps to mid 30bps, as shown on Chart 5).
While this might not appear that material, consider the alternative: for the Peruvian curve, spreads tightened by about 15bps during the same time periods (Chart 6). As a result, Peru’s spread premia above Chile basically was eliminated in the weeks following the protests.
More broadly, Chilean bonds underperformed the whole emerging market index, which was significant enough to be flagged by the media as well, with Bloomberg noting
The dollar notes due in 2050 and 2047 posted the first and third worst returns among all 74 emerging and frontier markets in the Bloomberg Barclays Emerging Markets Sovereign Index.
Social issues drove deterioration in asset fundamentals
Once again, we have observed that it can be hard to separate traditional evaluations from ESG ones, where some of Chile’s underperformance reflects the deterioration of traditional credit metrics resulting from the protests. But here as well, we think that is the point – social issues drove a deterioration of more typical fundamentals that in turn led to sovereign asset underperformance. The fact they are intertwined elevates the ESG factors in importance, and in our view shows the need to increasingly incorporate governance, inequality and institutional quality as part of sovereign credit analysis.
Investors reacted slightly differently here compared to the Brazil scenario. While Chile’s unrest forced them to give up hosting the COP 25 United Nations Climate Change Conference in December, there were not as many high profile investor communications or withdrawals from the region as with Brazil (though there were some).
Instead, Chile’s strong history of fiscal prudence, a reflection of their still strong governance, helped the country get through this crisis. As we noted in an article from October 2019:
Chile’s strong balance sheets, especially very low debt levels allow plenty of room for … additional debt (also at very attractive yields). In terms of the impact on economic activity, the medium-term impact should be modest, but given that they affect mostly the poor (basic pensions, minimum income), the fiscal multiplier should be rather high.
Thus the Chile example is one that shows both how social issues can influence asset prices, and how a history of strong governance can help a sovereign weather a crisis.
Three key takeaways that highlight some wider conclusions for ESG investing
As we have seen, in just these two case studies alone, there are a myriad of empirical examples showing how critical E, S and G issues can be in the pricing and value of assets — and there are plenty more case studies like this. Evidently, ESG issues and asset fundamentals are intrinsically linked.
Aside from the specific takeaways evident from each case study, there are also some broader conclusions we have found:
- It’s difficult to disentangle the direct effect of the ESG factors from the impact that those in turn have on traditional macro indicators. This is especially the case in Chile, where widespread protests/violence had negative effects on growth, the fiscal balance, etc. However, we think those very facts underscore the importance of incorporating ESG factors into traditional sovereign credit analysis. This is a clear example of the ‘S’ in Chile having noticeable, and in this case deleterious, impacts on the sovereign credit as a whole.
- In the case of Brazil, an argument can be made that the pressure on sovereign spreads seen in the aftermath of the publicising of the acceleration in deforestation could have been part of the reason why the government reversed some of the policies regarding the environment. While it is impossible to specifically prove this motivation and its exact contribution, we believe subjective evidence is there. If so, this would have very important implications for the overall role of ESG investing, and for the dialogue between policy-makers and investors to shaping mutually beneficial agendas around ESG.
- At face value, these case studies also suggest a potentially bigger impact on lower quality credits (Brazil impacted more than Chile), as well as greater weight given to environmental than social issues, at least in terms of disinvesting decisions by some global funds.
Clearly the ESG landscape and its effect on the investment environment moves and evolves quickly, especially in response to events that have a global impact. It is therefore impossible to ignore the potential ramifications that the global coronavirus crisis will have in this area. We expect that the market reaction and response could heighten the importance of ESG issues, in particular Social and Governance issues, more than ever. One can easily envision sovereign slippage on social or institutional considerations under the auspices of the virus, appropriate or not, which could have an impact on a sovereigns valuations moving forward. Conversely, we are already witnessing more cohesive responses in countries with better institutions. We look forward to discovering the new trends and behaviours that emerge.