[#17] Damages in the Dark

Putting numbers to climate damage in the latter half of this century is a mug’s game. I’ve argued extensively against the dominant economic models (DICE/RICE[1]) that pretend to real estimates. I’ve avoided the lure of quantified analysis because its pretence to precision is illusory, unwarranted, unreasonable and a dangerous distraction in policy debates. Qualified descriptions, while seemingly vague, better capture truth conditions of climate risk: catastrophic failure, threat to civic infrastructure and existential risk. The military and security establishments speak of threat multipliers and stresses to social fabric. These phrases resonate meaningfully. While I maintain plain language best captures what’s coming, recent work by actuaries - those most quantified of quantifiers! – begins to do that language justice.

?

Actuaries underwrite long-term financial activity by analyzing data to predict risk. Many were among the brightest of my applied mathematics peers at the University of Waterloo decades ago. Buried in language that bridged probability theory with finance, engineering and physical modelling with risk analysis, they went to work for large insurance and finance firms. The best among them generally become Chief Risk Officers, whose job it is to look to the horizon and back calculate probabilistic functions of financial risk. Whew. Basically, they apply models built on historical data to predict what’s going to happen. How many people die between the ages of 60 and 67, and how should we price life insurance? What flood claims are coming three decades out in British Columbia?

?

The professional challenge posed by climate risk to actuaries is profound indeed. Recently, one leading group - The Institute of and Faculty of Actuaries (IFoA) at the University of Exeter – addressed it. And to my mind, gets it about right. Their report - The Emperor’s New Climate Scenarios: Limitations and assumptions of commonly used climate-change scenarios in financial services – combines a refreshingly stark appraisal of prior efforts to quantify climate risk with a realistic assumption we’re headed to a hot-house world (3C+): “It is concerning to see...implausible results...that show benign, or even positive, economic outcomes in a hot-house world. This jars with climate science, which shows our economy may not exist at all if we do not mitigate climate change.”

?

The criticism of implausibility is key: dominant economic models (mainly DICE/RICE) have excluded from possibility the very catastrophic outcomes and planetary tipping points we are most worried about. Hence, their absurd outcome - which can be summarized “what, me worry?” For decades we’ve been lulled into complacence under the pretence of economic bull-artistry. It’s absurd to exclude monsters from a monster story, just because they’re scary. The actuary profession’s historical commitment to reliable advice[2] based on physical risk models, rather than economic abstractions, enabled (forced?) them to face these deep uncertainties head-on.

?

The paper itself is full of common-sense recommendations and observations (each of which got several pages in Frog): climate models underestimate risk; the remaining carbon budget is smaller than we think; group think lets crappy, but comforting, models dominate; time is too short to wait for perfect models; inter-disciplinary approaches are required to link the physical, economic and social aspects of climate risk; there’s no historical data to rely on as we’re entering uncharted territory; underestimated risk is a mistake that’s both catastrophic and unfixable; climate tipping points are uncertain gamechangers that swamp predictive outcomes. And so on. It’s a great paper.

?

The punchline is refreshing, and a doozy. We need to completely replace the dominant quadratic ‘damage function’ of traditional models. Those try, absurdly, to capture climate damage by guessing (there are no ‘estimates’ worthy of the name) what percentage hit climate brings to a ever-growing economy based on what little warming we’ve already seen. That guess then gets discounted a half century from that ever-growing economy to nearly nothing today. Same function applies whether we end up at 3C, 5C or even 10C. At 10C, everything’s fine! Small hit to long-term returns, but civilization and the economy just rumble on. That damage function is absurd, and Nordhaus himself knows it.

?

Instead, the IFoA starts with what we want to avoid and working back from there – it’s called a reverse stress test, which is used a lot in real-world financial analyses. That means acknowledging there is some point at which civilization effectively collapses and economic damage is complete, 100%. Whether you think that’s at 10C, 5C or 3C doesn’t matter - pick your poison. Then work backward along a logistic function – which looks like a crushed S curve – that plots economic damage against temperature.

?

Suddenly, we face a deeply scary and uncertain economic future. Everything is at stake. Which happens to match how the military talks about climate. And the deep intuitions of climate scientists. And those of people like me.

[1] Dynamic Integrated Model of Climate and the Economy and Regional Integrated Model of Climate and the Economy

[2] Their professional commitment to a ‘reliability objective’ helps: “To allow the intended user to place a high degree of reliance on actuarial information, practitioners must ensure the actuarial information, including the communication of any inherent uncertainty, is relevant, based on transparent and appropriate assumptions, complete and comprehensible.”