Residents flee as winds whip flames from the Morgan fire along Morgan Territory Road near Clayton, California, September 9, 2013.
Residents flee as winds whip flames from the Morgan fire along Morgan Territory Road near Clayton, California, September 9, 2013.
Noah Berger / Reuters

Exactly how bad is climate change going to be? That’s no small question. It’s also the wrong one. It’s precisely the uncertainty—the unknowns and perhaps unknowables—that should really be driving action to curb greenhouse-gas emissions today.

Climate science has made significant strides over the past few decades. Based on a plethora of information and observations, the confidence of the Intergovernmental Panel on Climate Change (IPCC) that global warming is being caused by human activity has gone from “more likely than not” in 1995 to “likely” in 2001 to “very likely” in 2007 to “extremely likely” in 2013. This is all the more striking, given that, by professional preference, scientists are wary of coming to definitive conclusions. By now, however, denying the reality of human attribution amounts to nothing short of willful blindness.

It is also impossible to ignore the issue that the final temperature rise—exactly how bad it’s going to get—is not certain. Of course, there’s no doubt that global temperatures have already warmed by around 0.85°C (1.5°F). The increase has been at least twice as high at the poles: bad news for ice caps and beachfront properties alike. Arctic sea ice has already lost half its area and three-quarters of its volume in the past three decades alone. The Foreign Affairs article “The Coming Arctic Boom” takes all of that as a given. Global average sea levels have been increasing at faster rates, leading to average projections of one to three feet by century’s end.

But global warming doesn’t stop in 2100. That’s where the all-important climate sensitivity parameter comes in: what happens to eventual temperatures as concentrations of greenhouse gases in the atmosphere double. The verdict after decades of research and tens of thousands of model runs: a doubling of concentrations is expected to lead to “likely” warming of between 1.5° and 4.5°C (2.7° and 8°F). The lower end is bad enough. The upper end is far worse.

Here’s where it gets tricky: despite amazing advances in climate science, we’ve now been looking at this same range for 35 years. We have more confidence today that the range is correct, but we haven’t been able to narrow it. In 2007, the IPCC tried. Research at the time led them to conclude that the lower bound of 1.5° (2.7°F) seemed unduly optimistic. IPCC consensus narrowed the range to between 2° and 4.5°C (3.6° and 8°F). By 2013, however, new research prompted the IPCC to widen its range once again to include 1.5° (2.7°F). Cue the headline: “Global Warming Not as Bad as Feared”! If only.

Graffiti on a wall next to the Regent's Canal, in Camden in London, December 22, 2009.
Graffiti on a wall next to the Regent's Canal, in Camden in London, December 22, 2009. 
Luke MacGregor / Reuters

In fact, far from being good news, the decision simply showed that climate sensitivity is indeed more persistently uncertain than previously thought possible. And basic economics—and human psychology—tells us that such risks are unwelcome.

Take a simple thought experiment. In one world, we know with certainty that climate sensitivity is 3°C (5.4°F). In the other, we know it could either be 1.5° or 4.5°C (2.7° or 8°F). The average would still be 3°C (5.4°F), but now there’s a chance that things may turn out better than feared, or worse. Fundamental human nature favors the first world, with a certain shot at 3°C (5.4°F), over the second.

Think of the stock market, where equity evaluations are so uncertain as to command outsized risk premiums. It’s often the unexplained and perhaps unexplainable high-impact, low-probability events that dominate the final outcome expressed in the risk premium. There’s a good reason why each weekday has been at least once called “black”: from Black Monday, on October 19, 1987, to Black Friday, on September 24, 1869—not to be confused with the Black Week, beginning Monday, October 6, 2008, which rang in the Great Recession.

The same reasoning applies in the climate world. Deep uncertainties prevail, and long-term temperature forecasts are only one aspect. Last time the concentration of carbon dioxide was as high as it is in the atmosphere today—400 parts per million (ppm)—was during the Pliocene epoch, over three million years ago. Global average temperatures were about 1°–2.5°C (1.8°–4.5°F) higher than they are today. Sea levels were up to 20 meters (66 feet) higher. Camels roamed in what is now Canada.

The best available climate models come close in their temperature projections to what the world experienced during the Pliocene, but they aren’t predicting sea levels of 20 meters (66 feet) higher. Nor do they predict camels wandering around Canada. Not now, and not hundreds of years from now.

Indian laborers carry small pieces of coal generated from a thermal power plant in Kolaghat 45 miles West of Calcutta, August 14,2002.
Indian laborers carry small pieces of coal generated from a thermal power plant in Kolaghat 45 miles West of Calcutta, August 14,2002. 
Jayanta Shaw / Reuters

That’s true for two important reasons. First, by necessity, climate models are often unduly skewed toward the known. Until recently, most climate models predicted rising sea levels only based on thermal expansion of the oceans (and the melting of mountain glaciers), but they did not include the effects of melting ice sheets. Call it a “known unknown.” Scientific understanding of melting polar ice caps had been so poor that most models simply left it out.

The second reason brings us back to climate sensitivity, the all-important link between carbon dioxide in the atmosphere and eventual temperature increases, and into the world of unknown unknowns. There are some things we simply don’t know, can’t know. And yes, most of those point in one direction.

We can sensibly cut off the lower bound of climate sensitivity. Adding carbon dioxide into the atmosphere doesn’t decrease temperatures. In fact, we can comfortably cut off the lower bound at something much closer to 1°C (1.8°F). Temperatures have already warmed by around 0.85°C (1.5°F), and we have increased atmospheric carbon dioxide concentrations by only around 40 percent above preindustrial levels. Nor are we close to the long-term equilibrium warming expected from the carbon dioxide already in the atmosphere. Temperatures and sea levels are bound to rise for decades and centuries, even if we stopped emitting carbon dioxide today.

We aren’t as lucky on the other end, where we can’t cut off things at 4.5°C (8°F), the upper bound of the IPCC’s likely range, or close to it. And it’s what happens in that upper tail of the distribution that may well define the final outcome. That’s where the true costs lie. It’s those supposed tail risks that have given us the black stock market days. The emphasis, of course, is on “supposed.” It turns out that these tail risks are, in fact, happening more often than feared. Climate change is no different.

So, yes, knowing precisely how bad it’s going to get is complicated. But no, that does not mean that we can just sit back and relax. Risks are pervasive, and they are not our friend. Gambles may be appealing in a casino, but in the real world humans crave certainty. That’s especially true when the stakes are so high.

We know that no matter the final temperature outcome, there is an urgent need to price the risk into our decision-making today. That means a price on carbon—estimated to be at least around $40 per ton of carbon dioxide emitted today. Any action to curb carbon dioxide that comes at a lower price is certainly worth the investment. Getting serious about risks only drives the value of action up further still.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now