Kevin Bourne, Head of Sustainable Finance at IHS Markit, says important steps are being taken to help institutional investors measure and manage climate-related risks, but many more remain.
Last November, the International Monetary Fund published a blog pointing out that stock markets were still not pricing climate risk accurately, despite US$1.3 trillion in direct damages from natural disasters over the previous decade. What more evidence do they need?, the blog seemed to shout.
As an equity trader turned Head of Sustainable Finance at leading data provider IHS Markit, Kevin Bourne knows exactly why the equity markets still struggle to price in climate risk.
First, long-term forecasting is not really part of the required skill set for the typical equity market participant. As head of electronic trading at Fidelity, Citi and HSBC, Bourne was used to dealing in milliseconds. For some, it was a challenge to think beyond the next quarterly results, let alone projecting 30 years into the future, to consider which business models might prosper in the net-zero future we’re now planning for. Insurers and actuaries understand how to price long-term risks, he says, not portfolio managers and equity traders.
“We are starting to understand how to price [a type of] risk that we’ve priced in insurance, but which we haven’t really priced in asset management or banking before, and that is forward-looking risk. Yes, the VIX is a forward-looking indicator, but, as a general rule, we like a nice historic time series,” says Bourne. “Now, investors are being explicitly asked to look out as far as 2050 and decide what these risks are on their portfolios.”
Second, even if the equity trader was minded to switch her attention to forward-looking scenarios and models, could she be certain of finding the data to populate them accurately?
Bourne insists we are still in the early stages of collecting the data and building the models needed to understand and calculate the three climate-related risks to financial stability cited by then-Bank of England Governor Mark Carney in his famous 2015 speech at Lloyds of London: physical risks; liability risks; and transition risks.
“I don’t know that we’ve necessarily unpacked those in enough detail yet as an industry. One of the reasons is that we don’t have enough data to start calculating what the risks could be,” Bourne observes.
Adapting at speed
The pace of change is a fundamental issue, with multiple factors in play. On the one hand, Bourne recognises the urgent need for asset owners and managers to base their investment decisions on accurate reported data about the climate risks impacting their portfolios, removing some of the assumptions present in today’s over-reliance on modelled data.
On the other, he understands the sheer scale and complexity of the challenge being taken on by regulators and market participants in rapidly assembling and analysing data in order to answer questions not previously widely contemplated.
As such, he is keen for the markets to avoid the situation he and colleagues found themselves during the 2008 global financial crisis, let down by flawed models which shared too many common assumptions.
“As an industry, we’re going through a very rapid structural and technical learning curve, at the same time that we’re going through a very complex and growing need to collect data that hasn’t been providing before, at the same time that regulations are changing very quickly, at the same time that investors are wanting answers,” says Bourne.
In terms of the accurate measurement and management of the physical risks from climate change, the finance sector is on the verge of a big step forward. For many, the steps required to transition from a range of voluntary sustainability reporting mechanisms to a system of regulatory disclosures, in parallel with the establishment of a new standards-setting body by the IFRS Foundation, can’t happen quickly enough.
Bourne says this shift is taking place “at a relatively fast pace by industry standards”, citing the scale of ambition evident in the Corporate Sustainability Reporting Directive announced by the European Commission last month. There is significant effort involved if the wide range of public and private firms covered by the new disclosure regime are to be ready to deliver the required data to a central facility by next October.
“The regulators understand that markets want to price this risk, but they can’t price the risk without effective data against which risk calculations can be undertaken,” says Bourne. “Over the next 5-7 years, most public companies generally and private companies of size globally are going to have to make ESG disclosures. Any delay in the market having access to that data keeps the point at which risk can be priced effectively further away.”
Although potentially painful for some, Bourne says the shift to greater mandatory disclosure is essential for accuracy. “You don’t have to go down very far into most ESG datasets to realise that you’re looking more at modelled data than you are disclosed data. You only have to go into the mid-1,000s of companies before the matrix starts to get sparse.”
Measuring the transition
For data vendors and their institutional clients, data from regulatorily-mandated disclosures is still only part of the picture. Bourne says an ever-widening range of sources are becoming available to help assess the physical risks from climate change. But there are big gaps to be filled in terms of the information needed to assess the liability and transition risks of climate change, too.
Bourne says there are important unanswered questions over the global capacity to offset liability risk – including the various risks from parties which have suffered loss or damage from the effects of climate change seeking compensation – whether through insurance or other forms of hedging.
However, transition risks are the hardest to assess, manage or provide data on. When many of the processes and assumptions that have driven industrial progress for 250 years are out of bounds, the difficulties of economic and business transition are mind-bending. Bourne suggests the structural changes required are beyond our current comprehension. From a risk point of view, he says, it is necessary to start by looking at individual companies’ goods products and services.
“Is the utility of the industrial output of Company A versus Company B such that one will or will not survive, because it’ll be either politically, legally or socially unacceptable? There are risks at different levels and all of them have capacity issues because, if you’ve got to change from business model A to B, you still have to borrow in order to work your way through that transition,” he says.
Translating the data
Clearly, the challenge of incorporating climate-related risk data – let alone other ESG metrics – into investment decision-making processes is multi-layered. Like many other data providers, IHS Markit is finding demand among institutional clients for raw data as well as scores which draw on underlying metrics, to come up with more easily digestible inputs.
Typically, a data model designed to create a single ESG score for a company or security starts off with raw ‘level 5’ data, which undergoes subsequent levels of abstraction. At level 4, raw data is converted into as many as 100-200 indicators. From these, vendors build their level 3 (sub-themes) and level 2 metrics (E, S or G), before reaching a ‘top of the pyramid’ single composite score.
Inevitably, differences in input and process lead to differences in outcome. Vendors collect different amounts of raw data, ask questions and convert written responses differently.
“The process of converting written into numerical information introduces bias. It’s a translation from qualitative into quantitative,” explains Bourne. “That’s a problem, but it’s a reaction to the fact that the [reporting] frameworks ask for a lot of written answers which have to be translated into numbers. You can’t hedge a paragraph.”
Further, in the event of missing or incomplete information, vendors may need to create a calculation engine to deal with a sparse matrix, creating synthetic scores. For each level of abstraction, they may apply a different weighting in their calculations.
Bourne accepts the frustration that can arise from the varying outcomes, in terms of the scores ascribed to companies or assets, and the growing attention of regulators. But, remembering 2008, he argues there can be strength in diversity.
“It’s a good thing the metrics are different,” he says. “After all, we know what happens when you clone calculation processes. But it does create a requirement for people to understand them, and to see where the signals are and how models have been created, rather than saying they’re wrong because they don’t look the same.”
Modelling risk
The increasing demand for raw data and the growing presence of quants are welcomed by Bourne as a sign that financial institutions are getting to grips with the task of modelling new risks. Among asset managers, it also reflects a need not only to comply with regulatory requirements, but also to position themselves commercially and meet evolving demand. Bourne draws parallels with the growing use of factor-based models by fund managers in the early 1990s.
“Everyone started off with two or three factors, but eventually managers wanted more granular data, because investors wanted to know: What’s your version of the truth? Are you just taking somebody else’s metrics and saying that’s okay?”
In response, IHS Markit is opening up its climate scenario modelling to interrogation and adaption by customers. The firm publishes three energy and climate scenarios, updated annually, which provide views of the future to 2050. Its ‘base case’ scenario, called Rivalry, is designed to provide a comprehensive outlook for planning purposes, while alternative scenarios, Autonomy and Discord, outline less optimal outcomes.
The three scenarios were most recently updated in July 2020 and none predict that the UN’s goal of limiting climate change to 1.5°C above pre-industrial levels will be met. They strongly reflect the impact of the pandemic, predicting cycles of relapse in the fight against the virus and reduced GDP and energy demand, as well as an increased role for the state, more automation and online communication, with public health fears and security concerns influencing political attitudes.
Less than a year later, some elements wear better than others. For good reason, for example, the sharp change of US climate policy under the Biden administration is not assumed. Crucially, says Bourne, clients are able to interact with the firm’s modelling platform to come up with their own scenarios.
“You can still get access to raw data as a customer, and therefore you can change what happens at a high level if you disagree with us,” he says.
Known unknowns
There is a long way to go, for regulators, investors and data providers alike, before climate risks are fully understood and priced in. Bourne is concerned that the scope European regulation is too ambitious, striving to provide a comprehensive response to demands for sustainable investments in short order, rather than focusing more narrowly on climate-related disclosures, at least in the short term.
In part, this worry stems from a recognition that overlaying climate and ESG risk scenarios and models onto existing asset/liability management frameworks is of itself a big ask for institutional investors such as pension funds. It also reflects the fact that the new models are far from mature, dealing as they do with new risks and new types of data.
“With whatever framework we build, how can we take into account all of the potential black swans out there? We know these things are going to happen in all probability, but we don’t know when, and, in some cases, we don’t know the scale. The impact on the economy of any of these occurring is really significant. If we’re already seeing positive feedback loops already in slow motion, what happens if one happens in fast motion?”
Despite these known unknowns, Bourne remains, like most equity traders, both a realist and an optimist.
“Humanity’s really good at surviving and adapting and mitigating. We can adapt to and mitigate against climate change, but we’ve got to remediate the damage of the past too. We’ve created the situation, and now we have to use our strengths to manage it.”
