The Price of a Degree: How Quantifying the Financial Ruin of 2°C is Changing Boardrooms

Climate value at risk is a term that quietly entered the financial lexicon but has since become a permanent fixture in the minds of central bankers, risk officers, and a growing number of corporate boards. It sounds like something from a scholarly article. However, it is now evident in stress tests, earnings calls, and the working papers distributed prior to significant investment decisions. Whether climate change is a risk to business is no longer the question being asked. The more difficult and significant question is precisely how much that risk is worth in dollars and when it begins to appear on the balance sheet.

Although the change has been developing for years, two relatively recent events contributed to its crystallization in a way that economists and regulators by themselves had not been able to. In 2015, Mark Carney, the Governor of the Bank of England at the time, gave a speech at Lloyd’s of London. Carney contended that in a world dedicated to keeping global warming to 2°C, fossil fuel companies might find themselves sitting on reserves that are, in his words, “literally unburnable.” The statement stuck because it was direct in a way that financial language seldom is. The second was Pacific Gas and Electric’s 2019 bankruptcy in California, which the Wall Street Journal called the first “climate-change bankruptcy.” It’s not a tale of financial mismanagement. Not a scandal involving derivatives. The physical realities of drought and wildfires, coupled with climate risk, caused a utility to collapse financially and legally.

The 2°C Financial Risk in Corporate Boardrooms: Key Facts

These were not hypothetical situations. These were actual occurrences, accompanied by share prices, court documents, and actual losses. And the way boardrooms viewed the discussion was altered by that tangibility and specificity.

What climate scenario analysis looks like in practice is less dramatic than either of those moments, but the implications run just as deep. Recently, a large food and beverage company working with PwC’s climate risk team needed to know how warming scenarios of 2°C and 4°C would impact the price of coffee and corn in its growing regions, as well as how carbon regulations might change the cost of plastic packaging. Physical mapping was the first step in the analysis, which used geospatial tools to forecast the frequency and intensity of climate hazards in particular agricultural regions. After that, it ran separate projections on how carbon pricing under the International Energy Agency’s net-zero pathway would affect plastics costs, used IPCC scenarios to model commodity price effects, and created a combined financial picture of the costs associated with various temperature outcomes across the company’s input supply chain. There was no qualitative warning as a result. The board could actually use the set of numbers.

When it released its recommendations in 2017, the Task Force on Climate-related Financial Disclosures was established to standardize this type of work across markets and industries. The architects of the framework had anticipated faster progress. The majority of businesses that disclose climate risks still do so in qualitative terms, characterizing their exposure’s nature without estimating its financial impact. Most businesses have yet to bridge the substantial gap between recognizing that supply chains may be impacted by flood risk and actually modeling the implications for earnings per share in a 2°C world. The IMF has been clear about this since joining the Network for Greening the Financial System in the years after Carney’s speech: before comprehensive climate stress testing becomes significant at scale, data gaps must be filled and standardized disclosure must be promoted.

There is a methodological component to the problem. Historical data, which is the foundation of most financial modeling, doesn’t reliably describe a climate future that has no historical precedent. Models that are calibrated against historical drought cycles, flood patterns, and hurricane frequencies are using inputs that the climate is actively rendering outdated. Businesses that use carbon pricing scenarios to test their business models are essentially conducting stress tests in future markets shaped by ongoing technological transitions against regulatory environments that do not yet exist. The accuracy of any quantitative output is constrained by this actual uncertainty. However, as PG&E showed, the alternative—treating climate as a qualitative footnote while physical and transition risks compound—has its own costs.

As the investment landscape changes, there’s a sense that the financial community has moved past the argument over whether climate risk actually affects balance sheets to a messier, more pragmatic one about how to measure it accurately enough to take action. The market for green bonds increased from $78 billion in 2015 to about $590 billion by 2019. Depending on the methodology used, estimates of sustainable finance assets worldwide now range from $3 trillion to $31 trillion. Boards are starting to link senior executive pay to performance goals related to climate change. Both the US Securities and Exchange Commission and the European Commission have proposed disclosure regulations that would require quantitative scenario analysis as standard procedure. Even when the destination is still unknown, the direction of travel is obvious.

Whether corporate adaptation is occurring at the same rate as physical risk is still up for debate. According to OECD estimates, the infrastructure investment needed to manage the climate transition will cost about $6 trillion annually through 2030. That figure is nearly impossible to comprehend, but it represents the true scope of what must be reconstructed, rerouted, and funded if the 2°C scenario is to remain a scenario rather than a definitive result. A beginning is being made by the boardrooms using climate value at risk models. The models themselves, with their ranges, assumptions, and inherent uncertainties, can describe but cannot provide an answer to the question of whether that start is adequate.